Compare commits

..

39 Commits

Author SHA1 Message Date
didericis c3f55357cc test(smolmachines): verify Linux sandbox escape gate + fix platform check
lint / lint (push) Successful in 1m56s
Empirical verification on NixOS/Linux with /dev/kvm + smolvm 1.4.7:

  1. DB path confirmed: ~/.local/share/smolvm/server/smolvm.db exists,
     schema matches (vms table, data BLOB with allowed_cidrs JSON field).

  2. smolvm 1.4.7 still silently drops --allow-cidr with --from
     (allowed_cidrs=None in DB after create). force_allowlist patch
     is still necessary and correct.

  3. All 5 sandbox-escape attacks blocked with BOT_BOTTLE_BACKEND=smolmachines
     on Linux: hostname/IP allowlist, HTTP DLP, DNS exfil, git-gate gitleaks.

Changes:
- test_sandbox_escape: remove darwin-only guard; allow linux too
- contrib/claude/Dockerfile: add dnsutils (dig) for attack 4 DNS test

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-09 00:49:04 -04:00
didericis 36e62cd7e8 test: add unit tests to pass diff-coverage gate (91%)
test / unit (pull_request) Successful in 57s
test / integration (pull_request) Successful in 16s
test / coverage (pull_request) Successful in 1m4s
lint / lint (push) Successful in 1m56s
prd-number / assign-numbers (push) Successful in 19s
test / unit (push) Successful in 1m0s
test / integration (push) Successful in 19s
test / coverage (push) Successful in 1m5s
Update Quality Badges / update-badges (push) Failing after 59s
Cover the new production lines added on this branch:

- cli/__init__: MissingEnvVarError maps to return 1
- cli/start: no-agents-defined path, prepare_with_preflight calls
  render_preflight, _text_render_preflight backend name output
- env: resolve_env raises MissingEnvVarError for missing interpolated var
- smolmachines/launch: _proxy_host Linux/non-Linux branches,
  _discover_urls git-gate host and supervise URL stamping

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-09 00:19:24 -04:00
didericis 70e58f1333 test: fix test_cache_miss and remove unused re import
lint / lint (push) Successful in 2m8s
test / unit (pull_request) Successful in 1m1s
test / integration (pull_request) Successful in 16s
test / coverage (pull_request) Failing after 1m6s
- test_smolmachines_launch_image: save mock now touches the tarball path
  so tarball.chmod(0o644) doesn't fail with FileNotFoundError
- test_smolmachines_launch (integration): remove unused `re` import
  (pyright reportUnusedImport)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-09 00:03:00 -04:00
didericis 3703b6f59f test: fix unit test failures after main rebase
lint / lint (push) Failing after 1m58s
test / unit (pull_request) Failing after 57s
test / integration (pull_request) Successful in 18s
test / coverage (pull_request) Failing after 56s
- test_egress: expect MissingEnvVarError instead of Die for missing/empty token refs
- test_git_gate, test_git_gate_render_provision: expect MissingEnvVarError instead of RuntimeError for missing forge token
- test_smolmachines_provision: handle 2 exec calls (mkdir -p + chmod/update-ca-certificates) in provision_ca
- test_smolmachines_smolvm: update machine create/delete argv assertions to --name flag form (smolvm 1.4.7+)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-08 23:46:47 -04:00
didericis afd943c4f9 fix(smolmachines): fix CA provisioning and git-http startup on Linux
lint / lint (push) Failing after 2m2s
test / unit (pull_request) Failing after 56s
test / integration (pull_request) Successful in 21s
test / coverage (pull_request) Failing after 53s
Two fixes for the Linux smolmachines backend:

- provision_ca: add mkdir -p before cp_in so the
  /usr/local/share/ca-certificates/ directory exists in the smolvm
  guest even when smolvm pack omits it (empty dir on Linux).
- git_http_backend: the fallback import in the except ImportError
  block tried `from git_gate import ...`, but git_gate.py is not
  deployed to the sidecar container and has its own package-relative
  imports; use the literal value 15 directly instead.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-08 22:06:28 -04:00
didericis 8f31741c9f refactor(errors): introduce MissingEnvVarError for missing host env vars
Replace die()/RuntimeError at the point of detection with a typed
MissingEnvVarError, and catch it in the CLI dispatcher alongside
ManifestError. Library code stays free of stderr output; the CLI
renders all user-facing error lines in one place.

Sites updated: egress.egress_resolve_token_values (unset/empty token),
git_gate._provision_dynamic_key and revoke_git_gate_provisioned_keys
(missing forge_token_env), env.resolve_env (unset interpolated var).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-08 22:05:49 -04:00
didericis 5bfe45bc1d fix(git-gate): use die() instead of RuntimeError for missing forge token env
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-08 22:05:03 -04:00
didericis 7aca7d5289 feat(start): show backend name in preflight output
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-08 22:01:58 -04:00
didericis f6d4eb9e3c fix(start): print error when no agents are defined
Previously start silently exited 0 when the manifest had no agents,
leaving the user with no feedback.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-08 22:01:58 -04:00
didericis c5b9f05467 fix(smolmachines): use bridge gateway as TSI proxy host on Linux
On Linux the guest kernel's LOCAL routing table routes all
127.0.0.0/8 to the guest's own loopback interface (priority 0,
checked before any main-table route), so TSI never sees
connections to the per-bottle loopback alias — the fix_guest_
loopback_routing approach confirmed this at the kernel level.

Use the per-bottle docker bridge gateway (192.168.N.1) instead.
It is not a loopback address, so the guest routes it via eth0
and TSI intercepts it normally.  The TSI allowlist remains a
/32 that is distinct from the container IP (192.168.N.2), so
direct bypass to egress:9099 is still blocked by TSI.

Changes:
- Add _proxy_host() helper: returns bundle_gateway on Linux,
  loopback alias on macOS
- Thread proxy_host through _start_bundle, _discover_urls,
  _launch_vm, and _bundle_launch_spec (publish_host_ip)
- Remove _fix_guest_loopback_routing (no longer needed)
- Relax proxy-URL assertion in the integration test to accept
  any http://IP:port (with a comment explaining the difference)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-08 22:01:58 -04:00
didericis-codex 09a2329f31 test(smolmachines): annotate pyright test helpers 2026-07-08 22:01:35 -04:00
didericis 03c89dae81 feat(smolmachines): run backend on Linux
Port the smolmachines backend so BOT_BOTTLE_BACKEND=smolmachines
works on Linux (KVM), not just macOS:

- Preflight gates /dev/kvm presence + accessibility on Linux with
  actionable remediation (kvm module, kvm group).
- smolvm state-DB path is platform-derived (XDG on Linux).
- force_allowlist runs on both platforms and is fail-closed: it
  verifies the persisted TSI allowlist and dies rather than booting
  a VM whose egress confinement it can't confirm. Previously it
  no-oped on Linux, failing OPEN.
- allocate() does per-bottle 127.0.0.<N> scoping on Linux too (no
  ifconfig needed — all of 127/8 is already loopback); only
  ensure_pool's lo0 aliasing stays macOS-only.
- README documents Linux + NixOS host setup.

Linux/KVM integration (the sandbox-escape acceptance gate) is
pending verification on a NixOS host; unit tests cover the new
platform branches.

Issue: #283

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NkwFXLFff9PYPy4wgVBJp9
2026-07-08 22:01:35 -04:00
didericis dd7232f368 docs(prd): add PRD for smolmachines backend on Linux
Design for porting the smolmachines backend off macOS-only: KVM
preflight, platform-aware smolvm state-DB path, fail-closed TSI
allowlist enforcement, and per-bottle loopback scoping on Linux.
NixOS is the primary validation target.

Issue: #283

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NkwFXLFff9PYPy4wgVBJp9
2026-07-08 22:01:35 -04:00
didericis 76fdefded3 fix: repair codex resume paths
lint / lint (push) Successful in 1m55s
test / unit (push) Successful in 57s
test / integration (push) Successful in 19s
test / coverage (push) Successful in 1m6s
Update Quality Badges / update-badges (push) Successful in 1m1s
2026-07-08 12:38:15 -04:00
didericis 1085a2280d fix(macos_container): mount supervise db's parent dir, not the file
lint / lint (push) Successful in 1m55s
test / unit (push) Successful in 59s
test / integration (push) Successful in 19s
test / coverage (push) Successful in 1m8s
Update Quality Badges / update-badges (push) Successful in 57s
Apple's `container run --mount type=bind` only accepts directory
sources — a file source fails with "path '<file>' is not a
directory". Move the sqlite db into its own ~/.bot-bottle/db/
subdirectory so it can be bind-mounted the same way the CA/routes
mounts already are, without exposing the rest of ~/.bot-bottle.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-07-06 15:48:36 -04:00
github-actions[bot] c97f4ecb8d ci(prd): assign sequential numbers to new PRDs 2026-07-06 19:29:43 +00:00
didericis-claude 142974a4b8 refactor: rename check_migrations → is_migrated
test / unit (pull_request) Successful in 58s
test / integration (pull_request) Successful in 18s
test / coverage (pull_request) Successful in 1m6s
lint / lint (push) Successful in 1m55s
prd-number / assign-numbers (push) Successful in 16s
test / unit (push) Successful in 54s
test / integration (push) Successful in 19s
test / coverage (push) Successful in 1m6s
Update Quality Badges / update-badges (push) Successful in 1m2s
Boolean-returning methods should read as predicates.
Renames DbStore.check_migrations, StoreManager.check_migrations,
and the test patch accordingly.
2026-07-06 19:19:49 +00:00
didericis-claude 041d9bbbf5 refactor: inline TableMigrations into store constructors
lint / lint (push) Successful in 1m59s
test / unit (pull_request) Successful in 57s
test / integration (pull_request) Successful in 20s
test / coverage (pull_request) Successful in 1m4s
Move _MIGRATIONS from module scope into AuditStore.__init__ and
QueueStore.__init__, treating table schema as a hidden implementation
detail of each store class.
2026-07-06 19:12:45 +00:00
didericis-claude 9fb83ef1b0 refactor: rename _supervise_types → supervise_types (treat as public)
lint / lint (push) Successful in 2m1s
test / unit (pull_request) Successful in 56s
test / integration (pull_request) Successful in 20s
test / coverage (pull_request) Successful in 1m6s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-06 18:53:53 +00:00
didericis-claude e7e8c7fdb4 refactor: extract supervise types to _supervise_types, eliminate get_supervise_mod()
Proposal, Response, AuditEntry, host_db_path, bot_bottle_root, and their
shared constants/helpers are moved to a new _supervise_types.py module.
queue_store and audit_store now import these directly instead of deferring
through get_supervise_mod() at call time.

host_db_path() in _supervise_types uses sys.modules[__name__].bot_bottle_root()
so monkey-patches on _supervise_types.bot_bottle_root propagate through it.
supervise.host_db_path() is re-defined to call bot_bottle_root() through
supervise's own globals, preserving patchability for callers that go through
the supervise module.

Test fixtures that create stores without an explicit db_path now patch
_sv_types.bot_bottle_root alongside supervise.bot_bottle_root. The server
test patches both flat and package _supervise_types since handle_tools_call
runs in the package context while the responder thread uses the flat module.

Dockerfile.sidecars gets a COPY for _supervise_types.py so the sidecar
bundle retains its flat import chain.

Follow-up to issuecomment-2872 on PR #320.
2026-07-06 18:53:25 +00:00
didericis-claude ed8fbfdfa8 feat: add StoreManager singleton for centralized store migration
lint / lint (push) Successful in 1m59s
test / unit (pull_request) Successful in 58s
test / integration (pull_request) Successful in 19s
test / coverage (pull_request) Failing after 1m3s
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-06 18:52:01 +00:00
didericis-claude 5fcca2060f feat(db_store): decouple migration from store construction
lint / lint (push) Failing after 2m4s
test / unit (pull_request) Successful in 1m2s
test / integration (pull_request) Successful in 18s
test / coverage (pull_request) Failing after 1m9s
Remove auto-migration from DbStore.__init__; add explicit check_migrations()
and migrate() methods. The CLI now checks both stores on every startup and
prompts the user to confirm before migrating. supervise.prepare() calls
.migrate() directly now that __init__ no longer does it implicitly.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-06 17:16:18 +00:00
didericis-claude a4d82e5ff2 refactor: extract TableMigrations and DbStore base class
lint / lint (push) Successful in 2m0s
test / unit (pull_request) Successful in 1m4s
test / integration (pull_request) Successful in 23s
test / coverage (pull_request) Successful in 1m9s
Adds bot_bottle/migrations.py (TableMigrations) and bot_bottle/db_store.py
(DbStore) per PR review. Both stores now inherit from DbStore and hold a
TableMigrations instance instead of duplicating schema-version logic inline.
2026-07-02 21:04:34 +00:00
didericis-claude e8e4f6f7c7 refactor: address PR review — rename, move helpers, add migration runner
lint / lint (push) Successful in 2m4s
test / unit (pull_request) Successful in 59s
test / integration (pull_request) Successful in 16s
test / coverage (pull_request) Successful in 1m5s
Per review #320 comments:

- Rename _sv() → get_supervise_mod() in both store files (review 206/211)
- Move _audit_entry_from_row onto AuditStore as _row_to_entry static method
  (review 208); move _proposal/_response_from_row onto QueueStore (review 211)
- Remove _host_db_path() free function; inline into __init__ (review 209/211)
- Add stdlib migration runner using a shared schema_versions table; each store
  tracks its own version under a module key so they can coexist in the same DB
  without clobbering a shared PRAGMA user_version (reviews 210/212/213)
- PRD: add goal 6 (migration runner), narrow non-goal to third-party ORM only
2026-07-02 03:27:02 +00:00
didericis-claude 5f0fc0d540 test: cover QueueStore/AuditStore guard branches and supervise bundle spec
lint / lint (push) Successful in 1m57s
test / unit (pull_request) Successful in 52s
test / integration (pull_request) Successful in 19s
test / coverage (pull_request) Successful in 1m2s
Add 8 tests covering the branches that were keeping diff-coverage below
90%: explicit db_path constructor arg, early-return guards when the DB
file is absent, _chmod OSError swallowing in both store classes, and the
supervise volume/env/daemon path in _bundle_launch_spec.

Diff-coverage rises from 89.2% to 94.6% (176/186 changed lines).
2026-07-02 02:59:26 +00:00
didericis-claude 244ad6a914 refactor: extract QueueStore and AuditStore to their own modules
lint / lint (push) Successful in 2m2s
test / unit (pull_request) Successful in 56s
test / integration (pull_request) Successful in 20s
test / coverage (pull_request) Failing after 59s
Moves _QueueStore → bot_bottle/queue_store.py (public QueueStore) and
_AuditStore → bot_bottle/audit_store.py (public AuditStore). Removes
the public queue_db_path() function; QueueStore resolves the DB path
via host_db_path() on the host, or via the SUPERVISE_DB_PATH env var
in the sidecar container (internal mechanism, not public API).

Adds queue_store.py and audit_store.py to Dockerfile.sidecars so the
sidecar bundle picks them up. Updates __all__ in supervise.py.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-07-01 21:45:08 +00:00
didericis-codex 29904609da fix(supervise): remove queue directory from db-backed flow
lint / lint (push) Successful in 2m4s
test / unit (pull_request) Successful in 59s
test / integration (pull_request) Successful in 20s
test / coverage (pull_request) Successful in 1m10s
2026-07-01 19:50:38 +00:00
didericis-codex 3067b067d2 fix(supervise): store queue rows in host sqlite db
lint / lint (push) Successful in 2m5s
test / unit (pull_request) Successful in 58s
test / integration (pull_request) Successful in 20s
test / coverage (pull_request) Successful in 1m2s
2026-07-01 19:33:43 +00:00
didericis-codex 212551df9a docs(prd): activate sqlite local storage
test / unit (pull_request) Successful in 55s
test / integration (pull_request) Successful in 18s
test / coverage (pull_request) Successful in 1m6s
2026-07-01 16:57:54 +00:00
didericis-codex f1b8bbdfa1 test(supervise): update edge cases for sqlite storage
lint / lint (push) Successful in 1m55s
test / unit (pull_request) Successful in 53s
test / integration (pull_request) Successful in 20s
test / coverage (pull_request) Successful in 1m10s
2026-07-01 16:57:45 +00:00
didericis-codex 08918f9a8a feat(supervise): store queue and audit data in sqlite
lint / lint (push) Failing after 1m53s
test / unit (pull_request) Failing after 45s
test / integration (pull_request) Successful in 17s
test / coverage (pull_request) Failing after 50s
2026-07-01 16:56:23 +00:00
didericis-codex 9af02831ea docs(prd): add sqlite local storage plan 2026-07-01 16:53:23 +00:00
Quality Badge Bot 5970b785aa chore: update quality badges
- Coverage: 83%
- Core coverage: 95%

[skip ci]
2026-07-01 16:51:08 +00:00
didericis 2f5cf81cf5 fix(git-gate): defer dynamic key provisioning
lint / lint (push) Successful in 1m59s
test / unit (push) Successful in 49s
test / integration (push) Successful in 23s
test / coverage (push) Successful in 1m0s
Update Quality Badges / update-badges (push) Successful in 53s
2026-07-01 12:45:46 -04:00
didericis 4a1e667306 fix(git-gate): inline GIT_GATE_TIMEOUT_SECS to fix git-http ImportError
lint / lint (push) Successful in 1m56s
test / unit (push) Successful in 48s
test / integration (push) Successful in 20s
test / coverage (push) Successful in 1m2s
Update Quality Badges / update-badges (push) Successful in 52s
git_http_backend.py is copied flat into the sidecar bundle image as a
standalone script, not as part of the bot_bottle package, and
git_gate.py/git_gate_render.py are never copied in. Its relative
import of GIT_GATE_TIMEOUT_SECS crashed the git-http daemon (port
9420) on every startup, silently leaving the smart-HTTP git-gate
transport down while the other sidecar daemons stayed up.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
2026-07-01 11:53:26 -04:00
didericis b93fe58523 feat(cli): add headless launch mode for orchestrators
test / unit (pull_request) Successful in 47s
test / integration (pull_request) Successful in 16s
test / coverage (pull_request) Successful in 1m4s
lint / lint (push) Successful in 2m0s
test / unit (push) Successful in 48s
test / integration (push) Successful in 18s
test / coverage (push) Successful in 57s
Update Quality Badges / update-badges (push) Successful in 57s
`--headless` is a non-interactive launch path for `cli.py start`:
agent, bottles, label, and color come from flags + manifest defaults
with no TUI selectors and no y/N preflight (auto-confirmed via a new
`assume_yes` param threaded into the shared `_launch_bottle` core).

- `--bottle` (repeatable) defaults to the agent's own `bottle:`;
  `--label` defaults to the agent name and auto-uniquifies on slug
  collision; `--color` defaults to none.
- `--prompt TEXT` is required in headless mode and is delivered to the
  agent via a new `headless_prompt(prompt)` method on `AgentProvider`,
  implemented for claude (`-p`), codex (positional), and pi (`-p`).
- The agent still execs on inherited stdio/PTY, so whatever allocates
  the PTY drives the live session; only the launch chrome is headless.
- `--headless --dry-run` previews the resolved plan without launching.

Adds unit coverage in tests/unit/test_cli_start_headless.py and
headless_prompt tests for each provider. Also stubs headless_prompt on
the in-test AgentProvider subclasses so the unit suite collects cleanly.

Closes #315.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WL77TgFxKbs3cidGMG9dz7
2026-06-30 15:08:14 -04:00
didericis 94eca35b4f fix(skills): validate skill names and quote provisioning paths
test / unit (push) Successful in 55s
test / integration (push) Successful in 23s
test / coverage (push) Successful in 1m11s
Update Quality Badges / update-badges (push) Successful in 1m3s
lint / lint (push) Successful in 2m18s
Skill names become host/guest path segments interpolated into the
`bottle.exec` shell strings in each contrib provider's provision_skills.
They were validated only as strings, so a name with shell metacharacters
or path traversal could reach the command.

Layer two defenses:
  - Primary: reject any skill name that isn't kebab-case
    ([a-z][a-z0-9-]*) at manifest load, reusing the convention already
    enforced on bottle/agent filenames (new is_valid_entity_name helper
    in manifest_schema). Fails loud and early, protecting every consumer
    of the name — not just the exec call sites.
  - Failsafe: shlex.quote the interpolated skills_dir / dst paths in the
    claude, codex, and pi providers, so a future unvalidated field can't
    inject shell metacharacters even if it bypasses the load-time check.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NkwFXLFff9PYPy4wgVBJp9
2026-06-27 02:15:30 -04:00
didericis f787764364 refactor(manifest): break import cycle by extracting ManifestBottle to a leaf module
test / unit (pull_request) Successful in 57s
test / integration (pull_request) Successful in 27s
test / coverage (pull_request) Successful in 1m23s
lint / lint (push) Successful in 2m24s
test / unit (push) Successful in 59s
test / integration (push) Successful in 26s
test / coverage (push) Successful in 1m17s
Update Quality Badges / update-badges (push) Successful in 1m13s
manifest.py imported the extends/loader resolvers, while those resolvers
needed ManifestBottle back from manifest.py — a true bidirectional cycle
papered over with in-function imports and TYPE_CHECKING guards (not clear
dependency inversion).

Extract ManifestBottle into a new leaf module manifest_bottle.py that depends
only on the other leaf modules (manifest_util/agent/egress/git/schema).
manifest.py re-exports ManifestBottle, so `from .manifest import ManifestBottle`
callers are unaffected. With the cycle gone:

- manifest_extends and manifest_loader import ManifestBottle from
  manifest_bottle and their other deps from the real source modules, all at
  top level (TYPE_CHECKING block removed).
- manifest.py imports the extends/loader/schema/yaml_subset/log helpers at
  module top; all per-function lazy imports in the cluster are removed.

No behavior change; full unit suite green, pyright clean.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01NkwFXLFff9PYPy4wgVBJp9
2026-06-26 23:42:03 -04:00
didericis-claude a256e5762a Merge pull request 'DLP injection-check perf, bounded variant cache, dedup supervise schema' (#312) from dlp-supervise-quality-fixes into main
lint / lint (push) Successful in 2m22s
test / unit (push) Successful in 50s
test / integration (push) Successful in 18s
test / coverage (push) Successful in 1m2s
Update Quality Badges / update-badges (push) Successful in 1m9s
2026-06-26 23:30:16 -04:00
72 changed files with 3015 additions and 997 deletions
+8 -2
View File
@@ -18,7 +18,7 @@
# /git-gate-entrypoint.sh docker-cp'd at start time
# /git-gate/creds/* docker-cp'd at start time
# /git/* bare repos, populated at runtime
# /run/supervise/queue/ bind-mounted at run time
# /run/supervise/bot-bottle.db bind-mounted at run time
# /home/mitmproxy/.mitmproxy/ mitmproxy CA dir
#
# Exposed ports inside the container:
@@ -66,6 +66,12 @@ COPY bot_bottle/egress_dlp_config.py /app/egress_dlp_config.py
COPY bot_bottle/egress_addon.py /app/egress_addon.py
COPY bot_bottle/dlp_detectors.py /app/dlp_detectors.py
COPY bot_bottle/yaml_subset.py /app/yaml_subset.py
COPY bot_bottle/migrations.py /app/migrations.py
COPY bot_bottle/db_store.py /app/db_store.py
COPY bot_bottle/supervise_types.py /app/supervise_types.py
COPY bot_bottle/queue_store.py /app/queue_store.py
COPY bot_bottle/audit_store.py /app/audit_store.py
COPY bot_bottle/store_manager.py /app/store_manager.py
COPY bot_bottle/supervise.py /app/supervise.py
COPY bot_bottle/supervise_server.py /app/supervise_server.py
COPY bot_bottle/sidecar_init.py /app/sidecar_init.py
@@ -81,7 +87,7 @@ RUN mkdir -p \
/etc/git-gate \
/git-gate/creds \
/git \
/run/supervise/queue \
/run/supervise \
/home/mitmproxy/.mitmproxy
# Documentation only — the compose renderer publishes whichever
+20 -4
View File
@@ -5,8 +5,8 @@
# bot-bottle
[![test](https://gitea.dideric.is/didericis/bot-bottle/actions/workflows/test.yml/badge.svg?branch=main)](https://gitea.dideric.is/didericis/bot-bottle/actions?workflow=test.yml)
[![coverage](https://img.shields.io/badge/coverage-84%25-brightgreen)](https://coverage.readthedocs.io/)
[![core coverage](https://img.shields.io/badge/core%20coverage-96%25-brightgreen)](https://gitea.dideric.is/didericis/bot-bottle/src/branch/main/docs/decisions/0004-coverage-policy.md)
[![coverage](https://img.shields.io/badge/coverage-83%25-brightgreen)](https://coverage.readthedocs.io/)
[![core coverage](https://img.shields.io/badge/core%20coverage-95%25-brightgreen)](https://gitea.dideric.is/didericis/bot-bottle/src/branch/main/docs/decisions/0004-coverage-policy.md)
**Problem:** Developer wants to run a coding agent without supervision, but they don't want a prompt injected or misbehaving agent wrecking their environment or exfiltrating sensitive data.
@@ -25,7 +25,7 @@
- **Provider templates (Claude, Codex)** — `Dockerfile.claude` / `Dockerfile.codex`, or a bottle-supplied Dockerfile. Claude auth via long-lived OAuth token; Codex via opt-in host device-auth forwarding.
- **gVisor auto-detect** — on Linux hosts where `runsc` is registered with Docker, every bottle launches under it for a userspace syscall barrier; no manifest config required.
- **Apple Container backend (macOS default when available)** — runs the agent and sidecar bundle with Apple's `container` CLI, using a host-only agent network plus a separate sidecar egress network.
- **Smolmachines backend** — runs the agent in a libkrun micro-VM while the sidecar bundle stays in Docker. TSI and smolmachines DNS filtering close the raw DNS exfiltration gap that exists in the legacy Docker backend.
- **Smolmachines backend** — runs the agent in a libkrun micro-VM while the sidecar bundle stays in Docker. TSI and smolmachines DNS filtering close the raw DNS exfiltration gap that exists in the legacy Docker backend. Runs on macOS (Hypervisor.framework) and Linux (KVM, `/dev/kvm`).
- **Legacy Docker backend** — still available for examples, CI, and hosts without Apple Container via `BOT_BOTTLE_BACKEND=docker` or `--backend=docker`.
## Architecture
@@ -71,10 +71,26 @@ When the agent exits, `cli.py` tears down every sidecar and both networks; nothi
## Quickstart
On compatible macOS hosts, the default backend requires Apple's `container` CLI and does not require Docker. The smolmachines backend requires Docker on the host for the sidecar bundle plus smolvm. The legacy Docker backend requires Docker. Claude bottles also need a long-lived Claude Code OAuth token (`claude setup-token`) exported as `BOT_BOTTLE_CLAUDE_OAUTH_TOKEN`.
On compatible macOS hosts, the default backend requires Apple's `container` CLI and does not require Docker. The smolmachines backend requires Docker on the host for the sidecar bundle plus `smolvm` (macOS or Linux). The legacy Docker backend requires Docker. Claude bottles also need a long-lived Claude Code OAuth token (`claude setup-token`) exported as `BOT_BOTTLE_CLAUDE_OAUTH_TOKEN`.
Use `BOT_BOTTLE_BACKEND=docker ./cli.py start <agent>` on hosts where Apple Container is not installed and Docker is the desired backend.
### smolmachines on Linux
The smolmachines backend runs on Linux as well as macOS. On Linux, `smolvm`/libkrun use KVM, so the host needs:
- **`/dev/kvm`** present and accessible. Load `kvm-intel` or `kvm-amd` (and enable virtualization in BIOS/firmware). The invoking user must be in the `kvm` group: `sudo usermod -aG kvm "$USER"` then re-login. bot-bottle preflights this and reports exactly what's missing.
- **`smolvm`** on `PATH`: `curl -sSL https://smolmachines.com/install.sh | sh`.
- **Docker** for the sidecar bundle and image build, same as macOS.
Per-bottle isolation works the same as macOS without any `ifconfig`/sudo step — all of `127.0.0.0/8` is already loopback on Linux, so each bottle's sidecar bundle is published on its own `127.0.0.<N>` and TSI's allowlist is scoped to that `/32`.
```sh
BOT_BOTTLE_BACKEND=smolmachines ./cli.py start <agent>
```
> **NixOS:** enable `virtualisation.docker`, ensure the KVM module is loaded (`boot.kernelModules = [ "kvm-intel" ];` or `kvm-amd`), and add your user to the `kvm` and `docker` groups. If you run bottles from a Gitea Actions runner, use a `host`-label runner so Docker, `smolvm`, and `/dev/kvm` are all reachable from the job. `smolvm` isn't in nixpkgs — install the release binary (pin the version) and put it on the runner's `PATH`.
```sh
./cli.py start <agent> # builds the image on first run, drops you into claude
```
+13
View File
@@ -209,6 +209,15 @@ class AgentProvider(ABC):
the supervise sidecar is reachable. No-op when
`plan.supervise_plan is None`."""
@abstractmethod
def headless_prompt(self, prompt: str) -> list[str]:
"""Return the agent CLI args that deliver `prompt` as the
initial task in a non-interactive (headless) session.
Called only when ``--prompt`` is passed to
``./cli.py start --headless``; the returned args are appended
after the provider's ``bypass_args`` and ``startup_args``."""
def provision_ca(self, bottle: "Bottle", plan: "BottlePlan") -> None:
"""Install the egress MITM CA into the agent's trust store.
@@ -218,6 +227,10 @@ class AgentProvider(ABC):
from .backend.util import AGENT_CA_PATH, log_ca_fingerprint, select_ca_cert
from .log import die
cert_host_path, label = select_ca_cert(plan.egress_plan)
# Ensure the target directory exists. smolvm's pack step may not
# preserve the empty /usr/local/share/ca-certificates/ directory
# on Linux; mkdir -p is idempotent and safe for all backends.
bottle.exec("mkdir -p /usr/local/share/ca-certificates", user="root")
bottle.cp_in(str(cert_host_path), AGENT_CA_PATH)
r = bottle.exec(
f"chmod 644 {AGENT_CA_PATH} && update-ca-certificates",
+91
View File
@@ -0,0 +1,91 @@
"""SQLite-backed audit store for supervise (PRD 0013)."""
from __future__ import annotations
import sqlite3
from pathlib import Path
try:
from .supervise_types import AuditEntry, host_db_path
from .db_store import DbStore
from .migrations import TableMigrations
except ImportError:
from supervise_types import AuditEntry, host_db_path # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
from db_store import DbStore # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
from migrations import TableMigrations # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
class AuditStore(DbStore):
"""SQLite-backed persistent store for supervise audit entries."""
def __init__(self, db_path: Path | None = None) -> None:
# One entry per schema version: migrations[0] brings a fresh DB to
# version 1, [1] to version 2, etc. Add new entries at the end; never
# edit existing ones.
migrations = TableMigrations("audit_store", [
# v1 — initial schema
"""
CREATE TABLE IF NOT EXISTS supervise_audit_entries (
id INTEGER PRIMARY KEY AUTOINCREMENT,
timestamp TEXT NOT NULL,
bottle_slug TEXT NOT NULL,
component TEXT NOT NULL,
operator_action TEXT NOT NULL,
operator_notes TEXT NOT NULL,
justification TEXT NOT NULL,
diff TEXT NOT NULL
)
""",
])
super().__init__(db_path or host_db_path(), migrations)
def write_audit_entry(self, entry: AuditEntry) -> Path:
with self._connect() as conn:
conn.execute(
"""
INSERT INTO supervise_audit_entries (
timestamp, bottle_slug, component, operator_action,
operator_notes, justification, diff
) VALUES (?, ?, ?, ?, ?, ?, ?)
""",
(
entry.timestamp,
entry.bottle_slug,
entry.component,
entry.operator_action,
entry.operator_notes,
entry.justification,
entry.diff,
),
)
self._chmod()
return self.db_path
def read_audit_entries(self, component: str, slug: str) -> list[AuditEntry]:
if not self.db_path.is_file():
return []
with self._connect() as conn:
rows = conn.execute(
"""
SELECT * FROM supervise_audit_entries
WHERE component = ? AND bottle_slug = ?
ORDER BY id
""",
(component, slug),
).fetchall()
return [self._row_to_entry(row) for row in rows]
@staticmethod
def _row_to_entry(row: sqlite3.Row) -> AuditEntry:
return AuditEntry(
timestamp=row["timestamp"],
bottle_slug=row["bottle_slug"],
component=row["component"],
operator_action=row["operator_action"],
operator_notes=row["operator_notes"],
justification=row["justification"],
diff=row["diff"],
)
__all__ = ["AuditStore"]
+4 -5
View File
@@ -34,7 +34,7 @@ from ...egress import (
from ...git_gate import GIT_GATE_HOSTNAME
from ...log import die, warn
from ...supervise import (
QUEUE_DIR_IN_CONTAINER,
DB_PATH_IN_CONTAINER,
SUPERVISE_HOSTNAME,
SUPERVISE_PORT,
)
@@ -163,16 +163,15 @@ def _sidecar_bundle_service(plan: DockerBottlePlan) -> dict[str, Any]:
if sp is not None:
env += [
f"SUPERVISE_BOTTLE_SLUG={plan.slug}",
f"SUPERVISE_QUEUE_DIR={QUEUE_DIR_IN_CONTAINER}",
f"SUPERVISE_DB_PATH={DB_PATH_IN_CONTAINER}",
f"SUPERVISE_PORT={SUPERVISE_PORT}",
]
volumes.append({
"type": "bind",
"source": str(sp.queue_dir),
"target": QUEUE_DIR_IN_CONTAINER,
"source": str(sp.db_path),
"target": DB_PATH_IN_CONTAINER,
"read_only": False,
})
internal_aliases = [EGRESS_HOSTNAME]
if gp.upstreams:
internal_aliases.append(GIT_GATE_HOSTNAME)
+9 -1
View File
@@ -37,7 +37,10 @@ from pathlib import Path
from typing import Callable, Generator
from ...egress import egress_resolve_token_values
from ...git_gate import revoke_git_gate_provisioned_keys
from ...git_gate import (
provision_git_gate_dynamic_keys,
revoke_git_gate_provisioned_keys,
)
from ...log import info, warn
from . import network as network_mod
from . import util as docker_mod
@@ -118,6 +121,11 @@ def launch(
git_gate_plan = plan.git_gate_plan
if git_gate_plan.upstreams:
git_gate_plan = provision_git_gate_dynamic_keys(
plan.manifest.bottle,
git_gate_plan,
git_gate_state_dir(plan.slug),
)
git_gate_plan = dataclasses.replace(
git_gate_plan,
internal_network=internal_network,
+30 -4
View File
@@ -28,9 +28,12 @@ from ...egress import (
egress_resolve_token_values,
egress_sidecar_env_entries,
)
from ...git_gate import revoke_git_gate_provisioned_keys
from ...git_gate import (
provision_git_gate_dynamic_keys,
revoke_git_gate_provisioned_keys,
)
from ...log import die, info, warn
from ...supervise import QUEUE_DIR_IN_CONTAINER, SUPERVISE_PORT
from ...supervise import DB_PATH_IN_CONTAINER, SUPERVISE_PORT
from ...util import expand_tilde
from ..docker.egress import EGRESS_CA_IN_CONTAINER, EGRESS_PORT
from ..docker.git_gate import (
@@ -98,6 +101,8 @@ def launch(
egress_network = egress_network_name(plan.slug)
_create_networks(internal_network, egress_network, stack)
plan = _provision_git_gate_keys(plan)
sidecar_name = sidecar_container_name(plan.slug)
container_mod.force_remove_container(sidecar_name)
_start_sidecar_bundle(plan, sidecar_name, internal_network, egress_network)
@@ -241,6 +246,19 @@ def _stamp_agent_urls(
)
def _provision_git_gate_keys(
plan: MacosContainerBottlePlan,
) -> MacosContainerBottlePlan:
if not plan.git_gate_plan.upstreams:
return plan
git_gate_plan = provision_git_gate_dynamic_keys(
plan.manifest.bottle,
plan.git_gate_plan,
git_gate_state_dir(plan.slug),
)
return dataclasses.replace(plan, git_gate_plan=git_gate_plan)
def _stage_git_gate(plan: MacosContainerBottlePlan, sidecar_name: str) -> None:
gp = plan.git_gate_plan
if not gp.upstreams:
@@ -361,7 +379,7 @@ def _sidecar_env_entries(plan: MacosContainerBottlePlan) -> tuple[str, ...]:
if plan.supervise_plan is not None:
env += [
f"SUPERVISE_BOTTLE_SLUG={plan.slug}",
f"SUPERVISE_QUEUE_DIR={QUEUE_DIR_IN_CONTAINER}",
f"SUPERVISE_DB_PATH={DB_PATH_IN_CONTAINER}",
f"SUPERVISE_PORT={SUPERVISE_PORT}",
]
return tuple(env)
@@ -387,7 +405,15 @@ def _sidecar_mounts(
sp = plan.supervise_plan
if sp is not None:
mounts.append((str(sp.queue_dir), QUEUE_DIR_IN_CONTAINER, False))
# `container run --mount type=bind` only accepts directory
# sources (a file source fails with "is not a directory") —
# mount db_path's dedicated parent dir instead of the file
# itself, same as the CA/routes mounts above.
mounts.append((
str(sp.db_path.parent),
str(Path(DB_PATH_IN_CONTAINER).parent),
False,
))
return tuple(mounts)
+14
View File
@@ -47,10 +47,24 @@ _HOME_FOR = {
"root": "/root",
}
_DEFAULT_PATH_FOR = {
# Committed smolmachine snapshots are rebuilt from a rootfs tarball and
# lose Docker image ENV metadata. Restore the provider CLI path here so
# resumed Codex bottles can still find the per-user install.
"node": (
"/home/node/.local/bin:"
"/home/node/.codex/packages/standalone/current/bin:"
"/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
),
"root": "/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin",
}
def _env_assignments_for(user: str, env: Mapping[str, str]) -> list[str]:
home = _HOME_FOR.get(user, f"/home/{user}")
out = [f"HOME={home}", f"USER={user}"]
if "PATH" not in env:
out.append(f"PATH={_DEFAULT_PATH_FOR.get(user, _DEFAULT_PATH_FOR['root'])}")
for k, v in env.items():
out.append(f"{k}={v}")
return out
+99 -42
View File
@@ -17,6 +17,7 @@ from __future__ import annotations
import dataclasses
import os
import platform
from contextlib import ExitStack, contextmanager
from pathlib import Path
from typing import Callable, Generator
@@ -27,7 +28,7 @@ from ...egress import (
egress_resolve_token_values,
egress_sidecar_env_entries,
)
from ...supervise import QUEUE_DIR_IN_CONTAINER, SUPERVISE_PORT
from ...supervise import DB_PATH_IN_CONTAINER, SUPERVISE_PORT
from ...util import expand_tilde
from ..docker import util as docker_mod
from ..docker.egress import (
@@ -41,7 +42,10 @@ from ..docker.git_gate import (
GIT_GATE_ENTRYPOINT_IN_CONTAINER,
GIT_GATE_HOOK_IN_CONTAINER,
)
from ...git_gate import revoke_git_gate_provisioned_keys
from ...git_gate import (
provision_git_gate_dynamic_keys,
revoke_git_gate_provisioned_keys,
)
from ...log import info, warn
from ...bottle_state import (
egress_state_dir,
@@ -88,12 +92,13 @@ def launch(
try:
loopback_ip, network = _allocate_resources(plan, stack)
plan = _mint_certs(plan)
plan = _start_bundle(plan, network, loopback_ip, stack)
plan = _discover_urls(plan, loopback_ip)
proxy_host = _proxy_host(plan, loopback_ip)
plan = _start_bundle(plan, network, proxy_host, stack)
plan = _discover_urls(plan, proxy_host)
agent_from_path = _agent_from_path(plan)
_launch_vm(plan, agent_from_path, loopback_ip, stack)
_launch_vm(plan, agent_from_path, proxy_host, stack)
_init_vm(plan)
bottle = SmolmachinesBottle(
@@ -141,10 +146,12 @@ def _allocate_resources(
) -> tuple[str, str]:
"""Reserve a loopback alias and create the per-bottle docker bridge.
macOS only routes 127.0.0.1 by default; the per-bottle alias
scopes TSI's allowlist to this bottle's published ports so the
agent can't reach other bottles' or host services' ports on
loopback. No-op on Linux."""
The per-bottle alias scopes TSI's allowlist to this bottle's
published ports so the agent can't reach other bottles' or host
services' ports on loopback. On macOS `ensure_pool` first
sudo-aliases the pool on `lo0`; on Linux that's a no-op since
all of 127.0.0.0/8 is already loopback, but the per-bottle
allocation runs on both."""
_loopback.ensure_pool()
loopback_ip = _loopback.allocate(plan.slug)
network = _bundle.bundle_network_name(plan.slug)
@@ -169,12 +176,13 @@ def _mint_certs(plan: SmolmachinesBottlePlan) -> SmolmachinesBottlePlan:
def _start_bundle(
plan: SmolmachinesBottlePlan,
network: str,
loopback_ip: str,
proxy_host: str,
stack: ExitStack,
) -> SmolmachinesBottlePlan:
"""Build the BundleLaunchSpec, resolve token env, start the
sidecar bundle container, and register teardown."""
bundle_spec = _bundle_launch_spec(plan, network, loopback_ip)
plan = _provision_git_gate_keys(plan)
bundle_spec = _bundle_launch_spec(plan, network, proxy_host)
token_env = _resolve_token_env(plan, dict(os.environ))
_bundle.ensure_bundle_image(bundle_spec.image)
_bundle.start_bundle(bundle_spec, env={**os.environ, **token_env})
@@ -182,41 +190,55 @@ def _start_bundle(
return plan
def _provision_git_gate_keys(
plan: SmolmachinesBottlePlan,
) -> SmolmachinesBottlePlan:
if not plan.git_gate_plan.upstreams:
return plan
git_gate_plan = provision_git_gate_dynamic_keys(
plan.manifest.bottle,
plan.git_gate_plan,
git_gate_state_dir(plan.slug),
)
return dataclasses.replace(plan, git_gate_plan=git_gate_plan)
def _discover_urls(
plan: SmolmachinesBottlePlan,
loopback_ip: str,
proxy_host: str,
) -> SmolmachinesBottlePlan:
"""Discover host-side ports for published container ports and
return the plan with URLs + guest_env stamped in.
Docker container IPs (192.168.x.x in the daemon's bridge)
aren't reachable from the smolvm guest on macOS — TSI uses
macOS networking, and macOS sees the daemon's bridge via the
published-port loopback forward only.
`proxy_host` is the host IP that both TSI's allowlist and
docker's port-forward bindings are keyed to. On macOS it is the
per-bottle loopback alias; on Linux it is the per-bottle bridge
gateway (see `_proxy_host`). The agent dials the published port
on this IP for all bundle-hosted services.
NO_PROXY includes the per-bottle loopback alias so the
supervise + git-gate URLs bypass HTTPS_PROXY."""
NO_PROXY includes `proxy_host` so supervise + git-gate URLs
bypass HTTPS_PROXY."""
agent_facing_host_port = _bundle.bundle_host_port(
plan.slug, _EGRESS_PORT, host_ip=loopback_ip,
plan.slug, _EGRESS_PORT, host_ip=proxy_host,
)
agent_proxy_url = f"http://{loopback_ip}:{agent_facing_host_port}"
agent_proxy_url = f"http://{proxy_host}:{agent_facing_host_port}"
agent_git_gate_host = ""
if plan.git_gate_plan.upstreams:
git_gate_host_port = _bundle.bundle_host_port(
plan.slug, _GIT_HTTP_PORT, host_ip=loopback_ip,
plan.slug, _GIT_HTTP_PORT, host_ip=proxy_host,
)
agent_git_gate_host = f"{loopback_ip}:{git_gate_host_port}"
agent_git_gate_host = f"{proxy_host}:{git_gate_host_port}"
agent_supervise_url = ""
if plan.supervise_plan is not None:
supervise_host_port = _bundle.bundle_host_port(
plan.slug, _SUPERVISE_PORT, host_ip=loopback_ip,
plan.slug, _SUPERVISE_PORT, host_ip=proxy_host,
)
agent_supervise_url = f"http://{loopback_ip}:{supervise_host_port}/"
agent_supervise_url = f"http://{proxy_host}:{supervise_host_port}/"
existing_no_proxy = plan.guest_env.get("NO_PROXY", "localhost,127.0.0.1")
no_proxy = f"{existing_no_proxy},{loopback_ip}"
no_proxy = f"{existing_no_proxy},{proxy_host}"
guest_env = {
**plan.guest_env,
"HTTPS_PROXY": agent_proxy_url,
@@ -246,27 +268,32 @@ def _discover_urls(
def _launch_vm(
plan: SmolmachinesBottlePlan,
agent_from_path: Path,
loopback_ip: str,
proxy_host: str,
stack: ExitStack,
) -> None:
"""Create, patch, and start the smolvm VM; register teardown.
--allow-cidr is the per-bottle loopback alias so the guest can
only reach this bottle's bundle ports. force_allowlist patches
smolvm 0.8.0's silent-drop of --allow-cidr when combined with
--from. Smolfile isn't usable here — smolvm 0.8.0 makes --from
--allow-cidr is `proxy_host/32` — the per-bottle loopback alias
on macOS or the bridge gateway on Linux (see `_proxy_host`). This
ensures the guest can only reach bundle ports published on that IP,
not the container IP directly. force_allowlist confirms the
allowlist persisted (patching smolvm 0.8.0's silent-drop of
--allow-cidr when combined with --from) and fails closed if it
can't. Smolfile isn't usable here — smolvm 0.8.0 makes --from
and --smolfile mutually exclusive."""
tsi_cidr = f"{proxy_host}/32"
_smolvm.machine_create(
plan.machine_name,
from_path=agent_from_path,
allow_cidrs=[f"{loopback_ip}/32"],
allow_cidrs=[tsi_cidr],
env=plan.guest_env,
)
stack.callback(_smolvm.machine_delete, plan.machine_name)
# Workaround smolvm 0.8.0: `--allow-cidr` is silently dropped
# when combined with `--from`. Patch the persisted state DB
# before start so the booted VM's TSI actually enforces.
_loopback.force_allowlist(plan.machine_name, [f"{loopback_ip}/32"])
# Confirm the booted VM's TSI allowlist will actually enforce the
# /32 before start (smolvm 0.8.0 silently drops `--allow-cidr`
# with `--from`, so the persisted state DB is patched if needed).
# Fails closed if enforcement can't be confirmed.
_loopback.force_allowlist(plan.machine_name, [tsi_cidr])
_smolvm.machine_start(plan.machine_name)
stack.callback(_smolvm.machine_stop, plan.machine_name)
@@ -275,7 +302,9 @@ def _init_vm(plan: SmolmachinesBottlePlan) -> None:
"""Repair filesystem ownership and wait for exec channel readiness.
Ownership repair: smolvm's pack process remaps files to the host
invoker's uid (501 on macOS). /home/node must be node:node so
invoker's uid (e.g. 501 on macOS, 1000 on Linux). The chowns use
names not numbers so they're correct on either. /home/node must
be node:node so
Claude Code can write ~/.claude.json; /tmp + /var/tmp need root
mode 1777 so non-root processes can create per-uid scratch dirs.
All folded into one sh -c to avoid back-to-back exec calls
@@ -298,8 +327,29 @@ def _init_vm(plan: SmolmachinesBottlePlan) -> None:
_smolvm.wait_exec_ready(plan.machine_name)
def _proxy_host(plan: SmolmachinesBottlePlan, loopback_ip: str) -> str:
"""Return the host IP for TSI's allowlist and docker port-forward bindings.
On macOS, the per-bottle loopback alias (e.g. ``127.0.0.16``) works
because macOS's network stack lets TSI intercept 127.x.x.x connects
from the guest before they reach the host's own loopback.
On Linux, the guest kernel's LOCAL routing table routes all
``127.0.0.0/8`` to the guest's own loopback — those packets never
reach eth0 and TSI never sees them. Using the per-bottle bridge
gateway (e.g. ``192.168.N.1``) instead sidesteps the problem: it
is not a loopback address, so the guest routes it via eth0 and TSI
intercepts it normally. The TSI allowlist is ``gateway/32``, which
is distinct from the container IP (``192.168.N.2``), so the agent
still can't reach the egress daemon directly — TSI blocks any
connection to the container IP that isn't via the published port."""
if platform.system() == "Linux":
return plan.bundle_gateway
return loopback_ip
def _bundle_launch_spec(
plan: SmolmachinesBottlePlan, network: str, loopback_ip: str,
plan: SmolmachinesBottlePlan, network: str, proxy_host: str,
) -> _bundle.BundleLaunchSpec:
"""Build a BundleLaunchSpec from the resolved inner Plans.
@@ -352,14 +402,15 @@ def _bundle_launch_spec(
daemons.append("supervise")
env += [
f"SUPERVISE_BOTTLE_SLUG={plan.slug}",
f"SUPERVISE_QUEUE_DIR={QUEUE_DIR_IN_CONTAINER}",
f"SUPERVISE_DB_PATH={DB_PATH_IN_CONTAINER}",
f"SUPERVISE_PORT={SUPERVISE_PORT}",
]
volumes.append((str(sp.queue_dir), QUEUE_DIR_IN_CONTAINER, False))
volumes.append((str(sp.db_path), DB_PATH_IN_CONTAINER, False))
# Container ports the agent reaches from the smolvm guest —
# published on host loopback so the guest can dial via TSI +
# macOS networking. Egress is always the agent's HTTP/HTTPS proxy.
# published on `proxy_host` so the TSI allowlist and the docker
# port-forward bindings point at the same IP. Egress is always
# the agent's HTTP/HTTPS proxy.
ports_to_publish: list[int] = [_EGRESS_PORT]
if gp.upstreams:
ports_to_publish.append(_GIT_HTTP_PORT)
@@ -376,7 +427,7 @@ def _bundle_launch_spec(
environment=tuple(env),
volumes=tuple(volumes),
ports_to_publish=tuple(ports_to_publish),
publish_host_ip=loopback_ip,
publish_host_ip=proxy_host,
)
@@ -450,6 +501,12 @@ def _ensure_smolmachine(image_ref: str, *, dockerfile: str = "") -> Path:
return sidecar
tarball = _SMOLMACHINE_CACHE_DIR / f"{digest}.image.tar"
docker_mod.save(image_ref, str(tarball))
# On Linux, `docker save -o` writes the tarball with owner-only
# permissions (mode 600). The crane push container runs as UID
# 65532 (distroless nonroot) and can't read it through a bind
# mount unless world-read is set. The tarball is temporary and
# lives in ~/.cache, so 644 is safe.
tarball.chmod(0o644)
try:
with ephemeral_registry() as handle:
push_ref = f"{handle.push_endpoint}/bot-bottle:{digest}"
@@ -33,10 +33,13 @@ sudo-add the missing pool on first use per boot — the aliases
persist on `lo0` until reboot, so subsequent launches don't
prompt.
Linux native daemons share the host's network namespace; the
whole `127.0.0.0/8` is reachable by default and aliases are
unnecessary. The pool logic detects native-Linux and skips sudo
entirely; the DB patch is also gated on macOS.
On Linux the whole `127.0.0.0/8` is already routed to `lo`, so
docker can publish a bundle's ports directly on `127.0.0.<N>`
with no `ifconfig`/sudo step. `ensure_pool` is therefore a no-op
on Linux, but per-bottle alias *allocation* and the TSI allowlist
DB patch run on both platforms — the isolation property is
identical, it's just cheaper to set up on Linux. The state-DB
path differs per platform (see `_smolvm_db_path`).
Allocation is coordinated by inspecting running bundle
containers' published host IPs — each bottle's bundle owns the
@@ -47,6 +50,7 @@ from __future__ import annotations
import fcntl
import json
import os
import platform
import re
import sqlite3
@@ -57,20 +61,34 @@ from typing import Iterable
from ...log import die, info
# smolvm's persistent VM state on macOS — a SQLite DB whose `vms`
# table holds one JSON BLOB per machine. The Linux path is
# different, but smolmachines is macOS-only in v1 (PRD 0023) so
# we hard-code this. If the file moves under us we'll see a
# clear FileNotFoundError; not worth defensive cross-platform
# detection until the backend actually needs Linux.
_SMOLVM_DB_PATH = (
def _smolvm_db_path() -> Path:
"""smolvm's persistent VM state — a SQLite DB whose `vms` table
holds one JSON BLOB per machine. macOS stores it under
`Application Support`; Linux follows the XDG base-dir spec
(`$XDG_DATA_HOME`, default `~/.local/share`).
NOTE: the Linux location is inferred from smolvm's documented
`~/.local/share` install layout and must be confirmed against a
real Linux smolvm install. If it's wrong, `force_allowlist`'s
fail-closed check turns it into a clear launch-time error rather
than a silent escape."""
if platform.system() == "Darwin":
return (
Path.home()
/ "Library"
/ "Application Support"
/ "smolvm"
/ "server"
/ "smolvm.db"
)
)
xdg_data = os.environ.get("XDG_DATA_HOME")
base = Path(xdg_data) if xdg_data else Path.home() / ".local" / "share"
return base / "smolvm" / "server" / "smolvm.db"
# Resolved once at import: the host platform doesn't change within a
# process. Tests patch this attribute directly.
_SMOLVM_DB_PATH = _smolvm_db_path()
# Sixteen aliases by default. Tunable for hosts that want more
@@ -131,30 +149,64 @@ def ensure_pool() -> None:
def force_allowlist(machine_name: str, allowed_cidrs: list[str]) -> None:
"""Patch smolvm's persistent VM-state DB to set the machine's
`allowed_cidrs` to the given list. Workaround for smolvm
0.8.0's silent-drop of `--allow-cidr` when used with `--from`.
"""Ensure the machine's persisted TSI allowlist equals
`allowed_cidrs`, failing **closed** if that can't be confirmed.
Must run AFTER `smolvm machine create` (the row has to
exist) and BEFORE `smolvm machine start` (smolvm reads the
row on start; in-flight VMs don't pick up changes). Once
smolvm honors the CLI flag upstream this whole function is
redundant — flag-respecting create + remove this call from
launch.
Runs on both macOS and Linux. It exists because smolvm 0.8.0
silently drops `--allow-cidr` when combined with `--from`, so
the allowlist has to be written into smolvm's persistent state
DB before `machine start`. Rather than assume the flag was
dropped, we read the persisted row and only patch when it
doesn't already match — so a newer smolvm that honors the flag
is left untouched.
No-op on non-macOS — the DB path differs and the Linux
smolmachines code path isn't exercised in v1."""
if not _is_macos():
return
Must run AFTER `smolvm machine create` (the row has to exist)
and BEFORE `smolvm machine start` (smolvm reads the row on
start; in-flight VMs don't pick up changes).
Fail-closed: if the state DB is missing, the row is missing, or
the allowlist still doesn't match after patching, we `die()`
rather than boot a VM whose egress confinement we can't verify
— an unconfirmed allowlist is a sandbox-escape risk (the agent
VM could reach all of host loopback)."""
want = list(allowed_cidrs)
if not _SMOLVM_DB_PATH.is_file():
die(
f"smolvm state DB not found at {_SMOLVM_DB_PATH}. "
f"smolvm 0.8.0 expected? `smolvm --version` to check."
f"smolvm state DB not found at {_SMOLVM_DB_PATH}; cannot "
f"confirm the TSI allowlist is enforced. Refusing to launch "
f"(fail-closed). Check `smolvm --version` and the DB "
f"location for your platform."
)
con = sqlite3.connect(str(_SMOLVM_DB_PATH))
try:
cur = con.cursor()
row = cur.execute(
cfg = _read_machine_cfg(con, machine_name)
if cfg.get("allowed_cidrs") != want:
cfg["allowed_cidrs"] = want
# Write as BLOB (the column type smolvm uses) — passing a
# plain str makes sqlite store it as Text and smolvm then
# fails to read it.
con.execute(
"UPDATE vms SET data = ? WHERE name = ?",
(sqlite3.Binary(json.dumps(cfg).encode()), machine_name),
)
con.commit()
cfg = _read_machine_cfg(con, machine_name)
if cfg.get("allowed_cidrs") != want:
die(
f"could not enforce TSI allowlist {want!r} for machine "
f"{machine_name!r} (persisted value is "
f"{cfg.get('allowed_cidrs')!r}). Refusing to launch "
f"(fail-closed)."
)
finally:
con.close()
def _read_machine_cfg(con: sqlite3.Connection, machine_name: str) -> dict[str, object]:
"""Read + JSON-decode a machine's `data` BLOB from the smolvm
state DB. Dies (fail-closed) if the row is missing — the caller
can't confirm enforcement without it."""
row = con.execute(
"SELECT data FROM vms WHERE name = ?", (machine_name,),
).fetchone()
if row is None:
@@ -162,18 +214,7 @@ def force_allowlist(machine_name: str, allowed_cidrs: list[str]) -> None:
f"smolvm DB has no row for machine {machine_name!r}"
f"machine_create must run before force_allowlist."
)
cfg = json.loads(row[0])
cfg["allowed_cidrs"] = list(allowed_cidrs)
# Write as BLOB (the column type smolvm uses) — passing a
# plain str makes sqlite store it as Text and smolvm then
# fails to read it.
cur.execute(
"UPDATE vms SET data = ? WHERE name = ?",
(sqlite3.Binary(json.dumps(cfg).encode()), machine_name),
)
con.commit()
finally:
con.close()
return json.loads(row[0])
def allocate(_slug: str) -> str:
@@ -184,16 +225,17 @@ def allocate(_slug: str) -> str:
used (no on-disk reservation, allocation is purely
docker-state-driven).
On non-macOS the whole `127.0.0.0/8` is loopback by default;
`127.0.0.1` is fine to share and we skip the alias dance.
This still returns a deterministic address so launch.py's
callers don't have to branch on platform.
Runs on both platforms: the allocation logic (docker-state
inspection + the file lock) is platform-independent. macOS
needs `ensure_pool` to have aliased the addresses on `lo0`
first; on Linux all of `127.0.0.0/8` is already loopback, so
docker can publish on the chosen `127.0.0.<N>` with no setup.
Per-bottle scoping (so the agent can't reach other bottles' or
host services' loopback ports) therefore holds on both.
An exclusive file lock serialises concurrent calls so two
simultaneous launches don't read the same docker state and
claim the same alias."""
if not _is_macos():
return "127.0.0.1"
_ALLOC_LOCK_PATH.parent.mkdir(parents=True, exist_ok=True)
with open(_ALLOC_LOCK_PATH, "w", encoding="utf-8") as lf:
fcntl.flock(lf, fcntl.LOCK_EX)
+10 -15
View File
@@ -116,10 +116,9 @@ def machine_create(
allow_cidrs: Sequence[str] = (),
env: Mapping[str, str] | None = None,
) -> None:
"""`smolvm machine create NAME [--image IMG | --from PATH]
[--allow-cidr CIDR ...] [-e K=V ...]`. NAME is positional
(the CLI's exception to the `--name` pattern other
subcommands use).
"""`smolvm machine create --name NAME [--image IMG | --from PATH]
[--allow-cidr CIDR ...] [-e K=V ...]`. NAME is passed as
`--name` (smolvm 1.4.7+; earlier versions took it positionally).
`image` (registry ref like `alpine:latest`) and `from_path`
(a `.smolmachine` artifact) are mutually exclusive — one or
@@ -133,12 +132,10 @@ def machine_create(
result without the Smolfile complication.
`--net` is sent explicitly when `allow_cidrs` is non-empty.
smolvm 0.8.0's docs say `--allow-cidr` implies `--net`, but
empirically the implication only fires when no `--from` is
set — `--from PATH --allow-cidr X/32` silently produces a
machine with `network: false` and no routes in the guest, so
the agent can't reach the bundle's pinned IP."""
args: list[str] = ["machine", "create"]
`--allow-cidr` implies `--net` per the CLI help, but sending
`--net` explicitly is harmless and ensures the guest has
network access even if that implication changes across versions."""
args: list[str] = ["machine", "create", "--name", name]
if image is not None:
args += ["--image", image]
if from_path is not None:
@@ -150,7 +147,6 @@ def machine_create(
if env:
for k, v in env.items():
args += ["-e", f"{k}={v}"]
args.append(name)
_smolvm(*args)
@@ -182,10 +178,9 @@ def machine_stop(name: str) -> None:
def machine_delete(name: str) -> None:
"""`smolvm machine delete -f NAME`. NAME is positional. `-f`
skips the interactive confirmation — required for
non-interactive teardown."""
_smolvm("machine", "delete", "-f", name)
"""`smolvm machine delete --name NAME -f`. `-f` skips the
interactive confirmation — required for non-interactive teardown."""
_smolvm("machine", "delete", "--name", name, "-f")
def machine_exec(
+39 -7
View File
@@ -5,19 +5,28 @@ unit-tested without importing the docker subprocess paths."""
from __future__ import annotations
import hashlib
import os
import platform
import shutil
from ...log import die
# libkrun's Linux backend drives the guest through KVM, so the host
# must expose `/dev/kvm` and the invoking user must be able to open
# it. macOS uses Hypervisor.framework and needs no device node.
_KVM_DEVICE = "/dev/kvm"
def smolmachines_preflight() -> None:
"""Ensure `smolvm` is on PATH before the launch flow runs.
Called from `_resolve_plan`; gives the operator a clear
install pointer rather than a cryptic FileNotFoundError
later. `gvproxy` is no longer required — see the PRD's design
pivot section."""
if shutil.which("smolvm") is not None:
return
"""Ensure the host can run the smolmachines backend before the
launch flow starts. Called from `_resolve_plan`; surfaces a
clear, actionable error instead of a cryptic `smolvm` failure
deep in launch.
Checks `smolvm` is on PATH (both platforms) and, on Linux,
that `/dev/kvm` exists and is accessible. `gvproxy` is no
longer required — see the PRD's design pivot section."""
if shutil.which("smolvm") is None:
die(
"BOT_BOTTLE_BACKEND=smolmachines requires `smolvm` on "
"PATH. Install with: "
@@ -25,6 +34,29 @@ def smolmachines_preflight() -> None:
"To use the legacy Docker backend instead, set "
"BOT_BOTTLE_BACKEND=docker or pass --backend=docker."
)
if platform.system() == "Linux":
_preflight_kvm()
def _preflight_kvm() -> None:
"""Linux-only: libkrun needs `/dev/kvm`. Distinguish 'KVM not
enabled' from 'no permission' so the operator knows which to
fix."""
if not os.path.exists(_KVM_DEVICE):
die(
f"BOT_BOTTLE_BACKEND=smolmachines needs {_KVM_DEVICE} on "
"Linux but it is missing. Enable KVM: load the kvm-intel "
"or kvm-amd kernel module (and confirm virtualization is "
"enabled in BIOS/firmware). To use the legacy Docker "
"backend instead, set BOT_BOTTLE_BACKEND=docker."
)
if not os.access(_KVM_DEVICE, os.R_OK | os.W_OK):
die(
f"{_KVM_DEVICE} exists but is not readable/writable by the "
"current user. Add your user to the `kvm` group "
"(`sudo usermod -aG kvm \"$USER\"`) and re-login, or run "
"with access to the device."
)
def smolmachines_bundle_subnet(slug: str) -> tuple[str, str, str]:
+2 -3
View File
@@ -284,9 +284,8 @@ def git_gate_state_dir(identity: str) -> Path:
def supervise_state_dir(identity: str) -> Path:
"""State subdir reserved for supervise sidecar bind-mount sources.
The queue dir is intentionally NOT under here — it lives at
~/.bot-bottle/queue/<slug>/ alongside the audit logs, so it
survives state-dir cleanup."""
Runtime queue/audit rows live in the host-level bot-bottle SQLite
database, so they survive state-dir cleanup."""
return bottle_state_dir(identity) / _SUPERVISE_SUBDIR
+18 -2
View File
@@ -7,8 +7,10 @@ from __future__ import annotations
import sys
from ..errors import MissingEnvVarError
from ..log import Die, die, error
from ..manifest import ManifestError
from ..store_manager import StoreManager
from ._common import PROG
from . import list as _list_mod
from .cleanup import cmd_cleanup
@@ -74,11 +76,25 @@ def main(argv: list[str] | None = None) -> int:
if handler is None:
usage()
die(f"unknown command: {command}")
mgr = StoreManager.instance()
if not mgr.is_migrated():
sys.stderr.write("bot-bottle: database schema is out of date\n")
sys.stderr.write("Migrate now? [y/N] ")
sys.stderr.flush()
try:
answer = sys.stdin.readline().strip().lower()
except EOFError:
answer = ""
if answer != "y":
error("migration required — re-run and confirm to migrate")
return 1
mgr.migrate()
try:
return handler(rest) or 0
except MissingEnvVarError as e:
error(str(e))
return 1
except ManifestError as e:
# Manifest/config problems surface as a catchable exception;
# print the reason and exit non-zero (same UX die() used to give).
error(str(e))
return 1
except Die as e:
+153 -10
View File
@@ -2,6 +2,11 @@
interactive claude-code session. The container is torn down when the
session ends.
`--headless` selects a non-interactive launch (agent/bottles/label from
flags, no TUI selectors, no y/N prompt) for orchestrators,
CI, and webhook dispatch. The agent still execs on the inherited
stdio/PTY, so an orchestrator that allocates the PTY drives the session.
The launch core is shared with `cli.py resume <identity>` through
the private orchestrator `_launch_bottle`.
"""
@@ -16,7 +21,7 @@ import tempfile
from pathlib import Path
from typing import Callable
from ..agent_provider import runtime_for
from ..agent_provider import get_provider, runtime_for
from ..backend import (
Bottle,
BottleSpec,
@@ -31,7 +36,7 @@ from ..bottle_state import (
is_preserved,
mark_preserved,
)
from ..log import info
from ..log import info, die
from ..manifest import Manifest, ManifestIndex
from ._common import PROG, USER_CWD, read_tty_line
from . import tui
@@ -50,6 +55,39 @@ def cmd_start(argv: list[str]) -> int:
"or host auto-selection). Overrides the env var when set."
),
)
parser.add_argument(
"--headless",
action="store_true",
help=(
"non-interactive launch: take agent/bottles/label from flags, "
"skip all prompts. For orchestrators, CI, and webhooks."
),
)
parser.add_argument(
"--bottle",
action="append",
default=None,
metavar="NAME",
help=(
"bottle to compose, repeatable (order = merge order). In "
"--headless, defaults to the agent's own bottle when omitted."
),
)
parser.add_argument(
"--label",
default=None,
help="bottle label / terminal title (--headless default: agent name)",
)
parser.add_argument(
"--color",
default=None,
help="bottle color, one of the 16 ANSI color names (--headless default: none)",
)
parser.add_argument(
"--prompt",
default=None,
help="initial task prompt delivered to the agent (required with --headless)",
)
parser.add_argument(
"name",
nargs="?",
@@ -61,9 +99,22 @@ def cmd_start(argv: list[str]) -> int:
dry_run = args.dry_run or os.environ.get("BOT_BOTTLE_DRY_RUN") == "1"
manifest = ManifestIndex.resolve(USER_CWD)
backend_name: str | None = args.backend
if args.headless:
return _start_headless(
manifest, args, dry_run=dry_run, backend_name=backend_name
)
agent_name: str | None = args.name
if agent_name is None:
if not manifest.all_agent_names:
print(
"bot-bottle: no agents defined. "
"Add an agent to ~/.bot-bottle/agents/ or ./bot-bottle/agents/ to get started.",
file=sys.stderr,
)
return 1
agent_name = tui.filter_select(
manifest.all_agent_names,
title="Select agent",
@@ -71,8 +122,6 @@ def cmd_start(argv: list[str]) -> int:
if agent_name is None:
return 0
backend_name: str | None = args.backend
# Bottle multiselect: always show after agent selection so operators
# can compose bottles at launch time without editing agent manifests.
available_bottles = manifest.all_bottle_names
@@ -109,6 +158,83 @@ def cmd_start(argv: list[str]) -> int:
)
# --- Headless launch -----------------------------------------------------
def _start_headless(
manifest: ManifestIndex,
args: argparse.Namespace,
*,
dry_run: bool,
backend_name: str | None,
) -> int:
"""Non-interactive launch path for orchestrators / CI / webhooks.
Resolves agent, bottles, label, and color from flags + manifest
defaults instead of the TUI selectors, and auto-confirms the
preflight. Otherwise runs the same launch core as the interactive
path, so the agent still execs on the inherited stdio/PTY — an
orchestrator allocates that PTY and relays it to its
desktop/mobile clients."""
agent_name = args.name
if not agent_name:
die("--headless requires an agent name: ./cli.py start <agent> --headless")
manifest.require_agent(agent_name) # raises ManifestError if unknown
prompt = args.prompt
if not prompt:
die(
"--headless requires --prompt: "
"./cli.py start <agent> --headless --prompt 'Do the thing'"
)
if args.bottle:
bottle_names: tuple[str, ...] = tuple(args.bottle)
else:
default_bottle = _peek_agent_bottle(manifest, agent_name)
if not default_bottle:
die(
f"--headless: agent '{agent_name}' has no default bottle; "
f"pass one or more --bottle NAME"
)
bottle_names = (default_bottle,)
label = _uniquify_label_headless(args.label or agent_name)
spec = BottleSpec(
manifest=manifest,
agent_name=agent_name,
copy_cwd=args.cwd,
user_cwd=USER_CWD,
label=label,
color=args.color or "",
bottle_names=bottle_names,
)
return _launch_bottle(
spec,
dry_run=dry_run,
backend_name=backend_name,
assume_yes=True,
headless_prompt_text=prompt,
)
def _uniquify_label_headless(label: str) -> str:
"""Non-interactive analog of `_resolve_unique_label`: if the label's
slug collides with a running bottle, append -2, -3, … until free,
logging the chosen label. Orchestrators fire-and-forget many bottles,
so silently picking a free name beats erroring on every collision."""
active_slugs = {a.slug for a in enumerate_active_agents()}
if docker_mod.slugify(label) not in active_slugs:
return label
n = 2
while docker_mod.slugify(f"{label}-{n}") in active_slugs:
n += 1
chosen = f"{label}-{n}"
info(f"label '{label}' already in use; using '{chosen}'")
return chosen
# --- Launch helpers ------------------------------------------------------
@@ -116,7 +242,7 @@ def prepare_with_preflight(
spec: BottleSpec,
*,
stage_dir: Path,
render_preflight: Callable[[DockerBottlePlan], None],
render_preflight: Callable[[DockerBottlePlan, str], None],
prompt_yes: Callable[[], bool],
dry_run: bool = False,
backend_name: str | None = None,
@@ -137,7 +263,7 @@ def prepare_with_preflight(
plan = backend.prepare(spec, stage_dir=stage_dir)
identity = _identity_from_plan(plan)
render_preflight(plan)
render_preflight(plan, backend.name)
if dry_run:
info("dry-run requested; not starting container.")
@@ -264,8 +390,9 @@ def _text_prompt_yes() -> bool:
def _text_render_preflight():
def _render(plan: DockerBottlePlan) -> None:
def _render(plan: DockerBottlePlan, backend_name: str) -> None:
print(file=sys.stderr)
print(f"backend: {backend_name}", file=sys.stderr)
print(_manifest_to_yaml(plan.manifest), file=sys.stderr)
return _render
@@ -376,10 +503,19 @@ def _launch_bottle(
*,
dry_run: bool,
backend_name: str | None = None,
assume_yes: bool = False,
headless_prompt_text: str = "",
) -> int:
"""Shared launch core for `start` and `resume`. Builds the plan,
prints / dry-runs / prompts as appropriate, brings the bottle up,
attaches claude, and prints the resume hint on session end."""
attaches claude, and prints the resume hint on session end.
`assume_yes` skips the interactive y/N confirmation (headless /
orchestrator launches), where there is no human at the prompt.
`headless_prompt_text` is passed to the provider's `headless_prompt`
method and the resulting args are appended to startup_args so the
agent receives the initial task without interactive input."""
stage_dir = Path(tempfile.mkdtemp(prefix="bot-bottle-stage."))
identity = ""
try:
@@ -387,7 +523,7 @@ def _launch_bottle(
spec,
stage_dir=stage_dir,
render_preflight=_text_render_preflight(),
prompt_yes=_text_prompt_yes,
prompt_yes=(lambda: True) if assume_yes else _text_prompt_yes,
dry_run=dry_run,
backend_name=backend_name,
)
@@ -397,10 +533,17 @@ def _launch_bottle(
backend = get_bottle_backend(backend_name)
with backend.launch(plan) as bottle:
agent_provider_template = getattr(plan, "agent_provider_template", "claude")
extra_args: tuple[str, ...] = ()
if headless_prompt_text:
extra_args = tuple(
get_provider(agent_provider_template).headless_prompt(
headless_prompt_text
)
)
exit_code = attach_agent(
bottle,
agent_provider_template=agent_provider_template,
startup_args=plan.agent_provision.startup_args,
startup_args=plan.agent_provision.startup_args + extra_args,
)
info(
f"session ended (exit {exit_code}); "
+9 -16
View File
@@ -45,7 +45,7 @@ from ..supervise import (
TOOL_EGRESS_BLOCK,
TOOL_GITLEAKS_ALLOW,
TOOL_EGRESS_TOKEN_ALLOW,
list_pending_proposals,
list_all_pending_proposals,
render_diff,
write_audit_entry,
write_response,
@@ -63,10 +63,9 @@ _REPORT_ONLY_TOOLS: tuple[str, ...] = (TOOL_GITLEAKS_ALLOW, TOOL_EGRESS_TOKEN_AL
@dataclass(frozen=True)
class QueuedProposal:
"""A pending proposal plus the queue dir it was found in."""
"""A pending proposal from the supervise queue."""
proposal: Proposal
queue_dir: Path
# Errors any remediation engine may raise. Caught by the TUI key
@@ -86,16 +85,11 @@ def apply_routes_change(slug: str, content: str) -> tuple[str, str]:
def discover_pending() -> list[QueuedProposal]:
"""Walk ~/.bot-bottle/queue/* and collect pending proposals."""
queue_root = _supervise.bot_bottle_root() / "queue"
if not queue_root.is_dir():
return []
out: list[QueuedProposal] = []
for slug_dir in sorted(queue_root.iterdir()):
if not slug_dir.is_dir():
continue
for proposal in list_pending_proposals(slug_dir):
out.append(QueuedProposal(proposal=proposal, queue_dir=slug_dir))
"""Collect pending proposals across bottles."""
out = [
QueuedProposal(proposal=proposal)
for proposal in list_all_pending_proposals()
]
out.sort(key=lambda q: q.proposal.arrival_timestamp)
return out
@@ -118,7 +112,6 @@ def _detail_lines(
(f"tool: {p.tool}", 0),
(f"id: {p.id}", 0),
(f"arrived: {p.arrival_timestamp}", 0),
(f"queue: {qp.queue_dir}", 0),
("", 0),
("justification:", 0),
]
@@ -165,7 +158,7 @@ def approve(
notes=notes,
final_file=final_file,
)
write_response(qp.queue_dir, response)
write_response(qp.proposal.bottle_slug, response)
_write_audit(
qp, action=status, notes=notes,
diff_before=diff_before, diff_after=diff_after,
@@ -179,7 +172,7 @@ def reject(qp: QueuedProposal, *, reason: str) -> None:
notes=reason,
final_file=None,
)
write_response(qp.queue_dir, response)
write_response(qp.proposal.bottle_slug, response)
_write_audit(qp, action=STATUS_REJECTED, notes=reason, diff_before="", diff_after="")
+1 -1
View File
@@ -21,7 +21,7 @@ FROM node:22-slim
# to it) works against egress's bumped TLS without the agent needing
# local DNS.
RUN apt-get update \
&& apt-get install -y --no-install-recommends git ca-certificates curl ripgrep \
&& apt-get install -y --no-install-recommends git ca-certificates curl ripgrep iproute2 dnsutils \
&& rm -rf /var/lib/apt/lists/*
# App-specific deps. Python isn't required by claude-code itself
+10 -3
View File
@@ -217,7 +217,7 @@ class ClaudeAgentProvider(AgentProvider):
if not agent.skills:
return
skills_dir = _skills_dir(plan.guest_home)
bottle.exec(f"mkdir -p {skills_dir}", user="root")
bottle.exec(f"mkdir -p {shlex.quote(skills_dir)}", user="root")
for name in agent.skills:
src = host_skill_dir(name)
if not os.path.isdir(src):
@@ -227,9 +227,13 @@ class ClaudeAgentProvider(AgentProvider):
)
dst = f"{skills_dir}/{name}"
info(f"copying skill {name} into {bottle.name}:{dst}")
bottle.exec(f"rm -rf {dst} && mkdir -p {dst}", user="root")
# Defense in depth: skill names are validated kebab-case at
# manifest load, but quote the path so a future unvalidated
# field can't inject shell metacharacters here either.
dst_q = shlex.quote(dst)
bottle.exec(f"rm -rf {dst_q} && mkdir -p {dst_q}", user="root")
bottle.cp_in(f"{src}/.", f"{dst}/")
bottle.exec(f"chown -R node:node {dst}", user="root")
bottle.exec(f"chown -R node:node {dst_q}", user="root")
def provision_prompt(self, plan: "BottlePlan", bottle: "Bottle") -> str | None:
"""Copy the prompt file into the guest, fix ownership/mode.
@@ -309,6 +313,9 @@ class ClaudeAgentProvider(AgentProvider):
f"claude mcp add --scope user --transport http supervise {supervise_url}"
)
def headless_prompt(self, prompt: str) -> list[str]:
return ["-p", prompt]
def _exec(bottle: "Bottle", script: str, error: str) -> None:
result = bottle.exec(script, user="root")
+20 -6
View File
@@ -34,6 +34,12 @@ if TYPE_CHECKING:
_SUPERVISE_MCP_NAME = "supervise"
_CODEX_CLI = "/home/node/.codex/packages/standalone/current/bin/codex"
_CODEX_CLI_PATH = (
"/home/node/.local/bin:"
"/home/node/.codex/packages/standalone/current/bin:"
"/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
)
def _skills_dir(guest_home: str) -> str:
@@ -50,7 +56,7 @@ def _prompt_path(guest_home: str) -> str:
_RUNTIME = AgentProviderRuntime(
template="codex",
command="codex",
command=_CODEX_CLI,
image="bot-bottle-codex:latest",
prompt_mode="read_prompt_file",
bypass_args=("--dangerously-bypass-approvals-and-sandbox",),
@@ -145,7 +151,8 @@ class CodexAgentProvider(AgentProvider):
"env",
f"HOME={guest_home}",
f"CODEX_HOME={auth_dir}",
"codex", "login", "status",
f"PATH={_CODEX_CLI_PATH}",
_CODEX_CLI, "login", "status",
), (
"codex host credentials: dummy auth was copied into the "
"guest, but Codex did not accept it"
@@ -183,7 +190,7 @@ class CodexAgentProvider(AgentProvider):
if not agent.skills:
return
skills_dir = _skills_dir(plan.guest_home)
bottle.exec(f"mkdir -p {skills_dir}", user="root")
bottle.exec(f"mkdir -p {shlex.quote(skills_dir)}", user="root")
for name in agent.skills:
src = host_skill_dir(name)
if not os.path.isdir(src):
@@ -193,9 +200,13 @@ class CodexAgentProvider(AgentProvider):
)
dst = f"{skills_dir}/{name}"
info(f"copying skill {name} into {bottle.name}:{dst}")
bottle.exec(f"rm -rf {dst} && mkdir -p {dst}", user="root")
# Defense in depth: skill names are validated kebab-case at
# manifest load, but quote the path so a future unvalidated
# field can't inject shell metacharacters here either.
dst_q = shlex.quote(dst)
bottle.exec(f"rm -rf {dst_q} && mkdir -p {dst_q}", user="root")
bottle.cp_in(f"{src}/.", f"{dst}/")
bottle.exec(f"chown -R node:node {dst}", user="root")
bottle.exec(f"chown -R node:node {dst_q}", user="root")
def provision_prompt(self, plan: "BottlePlan", bottle: "Bottle") -> str | None:
"""Copy the prompt file into the guest, fix ownership/mode.
@@ -263,7 +274,7 @@ class CodexAgentProvider(AgentProvider):
return
info(f"registering supervise MCP server in agent codex config → {supervise_url}")
r = bottle.exec(
f"codex mcp add {_SUPERVISE_MCP_NAME} --url "
f"{shlex.quote(_CODEX_CLI)} mcp add {_SUPERVISE_MCP_NAME} --url "
f"{shlex.quote(supervise_url)}",
user="node",
)
@@ -275,6 +286,9 @@ class CodexAgentProvider(AgentProvider):
f"codex mcp add supervise --url {shlex.quote(supervise_url)}"
)
def headless_prompt(self, prompt: str) -> list[str]:
return [prompt]
def _exec(bottle: "Bottle", script: str, error: str) -> None:
result = bottle.exec(script, user="root")
+10 -3
View File
@@ -238,7 +238,7 @@ class PiAgentProvider(AgentProvider):
if not agent.skills:
return
skills_dir = _skills_dir(plan.guest_home)
bottle.exec(f"mkdir -p {skills_dir}", user="root")
bottle.exec(f"mkdir -p {shlex.quote(skills_dir)}", user="root")
for name in agent.skills:
src = host_skill_dir(name)
if not os.path.isdir(src):
@@ -248,9 +248,13 @@ class PiAgentProvider(AgentProvider):
)
dst = f"{skills_dir}/{name}"
info(f"copying skill {name} into {bottle.name}:{dst}")
bottle.exec(f"rm -rf {dst} && mkdir -p {dst}", user="root")
# Defense in depth: skill names are validated kebab-case at
# manifest load, but quote the path so a future unvalidated
# field can't inject shell metacharacters here either.
dst_q = shlex.quote(dst)
bottle.exec(f"rm -rf {dst_q} && mkdir -p {dst_q}", user="root")
bottle.cp_in(f"{src}/.", f"{dst}/")
bottle.exec(f"chown -R node:node {dst}", user="root")
bottle.exec(f"chown -R node:node {dst_q}", user="root")
def provision_prompt(self, plan: "BottlePlan", bottle: "Bottle") -> str | None:
prompt_path = _prompt_path(plan.guest_home)
@@ -311,6 +315,9 @@ class PiAgentProvider(AgentProvider):
) -> None:
del plan, bottle, supervise_url
def headless_prompt(self, prompt: str) -> list[str]:
return ["-p", prompt]
def _exec(bottle: "Bottle", script: str, error: str) -> None:
result = bottle.exec(script, user="root")
+59
View File
@@ -0,0 +1,59 @@
"""Shared SQLite-backed store base class for bot-bottle (PRD 0013)."""
from __future__ import annotations
import sqlite3
from pathlib import Path
try:
from .migrations import TableMigrations
except ImportError:
from migrations import TableMigrations # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
class DbVersionError(Exception):
"""Raised when the on-disk schema is behind the current migration list."""
class DbStore:
"""Base for SQLite-backed stores. Subclasses resolve db_path then call super().__init__."""
def __init__(self, db_path: Path, migrations: TableMigrations) -> None:
self.db_path = db_path
self._migrations = migrations
self.db_path.parent.mkdir(parents=True, exist_ok=True)
def _connect(self) -> sqlite3.Connection:
conn = sqlite3.connect(self.db_path)
conn.row_factory = sqlite3.Row
return conn
def is_migrated(self) -> bool:
"""Return True if the DB is fully up-to-date, False if migration is needed."""
if not self.db_path.exists():
return False
try:
with self._connect() as conn:
row = conn.execute(
"SELECT version FROM schema_versions WHERE module = ?",
(self._migrations.schema_key,),
).fetchone()
except sqlite3.OperationalError:
return False
version = row[0] if row else 0
return version == len(self._migrations.migrations)
def migrate(self) -> None:
"""Apply any pending migrations and set permissions on the DB file."""
with self._connect() as conn:
self._migrations.apply(conn)
self._chmod()
def _chmod(self) -> None:
try:
self.db_path.chmod(0o600)
except OSError:
pass
__all__ = ["DbStore", "DbVersionError"]
+7 -4
View File
@@ -23,6 +23,7 @@ from .egress_addon_core import (
PathMatch as CorePathMatch,
Route,
)
from .errors import MissingEnvVarError
from .log import die
if TYPE_CHECKING:
@@ -354,16 +355,18 @@ def egress_resolve_token_values(
for token_env, token_ref in token_env_map.items():
value = host_env.get(token_ref)
if value is None:
die(
raise MissingEnvVarError(
token_ref,
f"egress: host env var '{token_ref}' is unset. Set it "
f"before launching, or remove the corresponding auth block "
f"from bottle.egress.routes."
f"from bottle.egress.routes.",
)
if not value:
die(
raise MissingEnvVarError(
token_ref,
f"egress: host env var '{token_ref}' is empty. The "
f"egress will not inject an empty token; set it to "
f"the real value or remove the route's auth block."
f"the real value or remove the route's auth block.",
)
out[token_env] = value
return out
+6 -9
View File
@@ -79,14 +79,13 @@ class EgressAddon:
# only — a restart re-prompts. Mutated only from the asyncio loop that
# runs the addon hooks, so no lock is needed.
self.safe_tokens: set[str] = set()
self._supervise_queue_dir = os.environ.get("SUPERVISE_QUEUE_DIR", "").strip()
self._supervise_slug = os.environ.get("SUPERVISE_BOTTLE_SLUG", "").strip()
self._token_allow_timeout = _token_allow_timeout_from_env(os.environ)
self._reload(initial=True)
self._install_sighup()
def _supervise_available(self) -> bool:
return bool(self._supervise_queue_dir and self._supervise_slug)
return bool(self._supervise_slug)
def _reload(self, *, initial: bool = False) -> None:
try:
@@ -393,9 +392,8 @@ class EgressAddon:
justification=_TOKEN_ALLOW_JUSTIFICATION,
current_file_hash=_sv.sha256_hex(payload),
)
queue_dir = Path(self._supervise_queue_dir)
try:
_sv.write_proposal(queue_dir, proposal)
_sv.write_proposal(proposal)
except OSError as e:
sys.stderr.write(
f"egress: could not queue token-allow proposal: {e}; "
@@ -411,8 +409,8 @@ class EgressAddon:
**self._req_ctx(flow),
}) + "\n")
response = await self._await_token_response(queue_dir, proposal.id)
_sv.archive_proposal(queue_dir, proposal.id)
response = await self._await_token_response(proposal.id)
_sv.archive_proposal(self._supervise_slug, proposal.id)
if response is not None and response.status in (
_sv.STATUS_APPROVED, _sv.STATUS_MODIFIED,
@@ -439,16 +437,15 @@ class EgressAddon:
async def _await_token_response(
self,
queue_dir: Path,
proposal_id: str,
) -> "_sv.Response | None":
"""Poll the queue dir for the operator's response without blocking the
"""Poll the DB for the operator's response without blocking the
proxy event loop. Returns the Response, or None on timeout."""
loop = asyncio.get_running_loop()
deadline = loop.time() + self._token_allow_timeout
while True:
try:
return _sv.read_response(queue_dir, proposal_id)
return _sv.read_response(self._supervise_slug, proposal_id)
except (OSError, ValueError, KeyError):
# Not written yet, or a partial/malformed write — retry until
# the deadline, then fail closed.
+4 -2
View File
@@ -35,6 +35,7 @@ import re
import sys
from dataclasses import dataclass, field
from .errors import MissingEnvVarError
from .log import die
from .manifest import Manifest
@@ -136,9 +137,10 @@ def resolve_env(manifest: Manifest) -> ResolvedEnv:
host_var = env_entry_interpolated_from(raw)
host_value = os.environ.get(host_var, "")
if not host_value:
die(
raise MissingEnvVarError(
host_var,
f"env entry {name} is interpolated from ${host_var}, "
f"but ${host_var} is unset or empty in the host environment."
f"but ${host_var} is unset or empty in the host environment.",
)
forwarded[name] = host_value
else: # literal
+18
View File
@@ -0,0 +1,18 @@
"""bot-bottle application-level exception hierarchy.
Exceptions here are caught by the CLI dispatcher (``cli/__init__.py``)
and rendered as ``bot-bottle: error: `` lines same UX as ``die()``,
but without coupling library code to stderr output."""
from __future__ import annotations
class MissingEnvVarError(Exception):
"""Raised when a required host environment variable is unset or empty.
``var_name`` is the exact name of the missing variable so callers
can display it without re-parsing the message string."""
def __init__(self, var_name: str, message: str) -> None:
self.var_name = var_name
super().__init__(message)
+7 -11
View File
@@ -30,7 +30,6 @@ backend-specific and lives on concrete subclasses (see
from __future__ import annotations
import dataclasses
from abc import ABC
from dataclasses import dataclass
from pathlib import Path
@@ -53,6 +52,7 @@ from .git_gate_render import (
_gitconfig_validate_value,
)
from .git_gate_provision import (
provision_git_gate_dynamic_keys,
revoke_git_gate_provisioned_keys,
_provision_dynamic_key,
_resolve_identity_file,
@@ -82,6 +82,7 @@ class GitGatePlan:
class GitGate(ABC):
"""The per-agent git-gate. Encapsulates the host-side prepare
(upstream lift + entrypoint/hook render); the sidecar's
@@ -93,20 +94,14 @@ class GitGate(ABC):
entrypoint, pre-receive hook, and access-hook scripts (mode
600) under `stage_dir`. Pure host-side, no docker subprocess.
For `gitea` key entries, also generates and registers
a fresh deploy key via the forge API and writes the private key
+ key ID to `stage_dir`.
For `gitea` key entries, the returned upstream intentionally
has an empty identity file. Backend launch fills that in after
the operator confirms the preflight.
Returned plan is incomplete: the launch step must fill
`internal_network` / `egress_network` via `dataclasses.replace`
before passing the plan to `.start`."""
upstreams_list = list(git_gate_upstreams_for_bottle(bottle))
for i, entry in enumerate(bottle.git):
upstreams_list[i] = dataclasses.replace(
upstreams_list[i],
identity_file=_resolve_identity_file(entry, slug, stage_dir),
)
upstreams = tuple(upstreams_list)
upstreams = git_gate_upstreams_for_bottle(bottle)
entrypoint = stage_dir / "git_gate_entrypoint.sh"
entrypoint.write_text(git_gate_render_entrypoint(upstreams))
entrypoint.chmod(0o600)
@@ -162,6 +157,7 @@ __all__ = [
"git_gate_render_entrypoint",
"git_gate_render_hook",
"git_gate_render_access_hook",
"provision_git_gate_dynamic_keys",
"revoke_git_gate_provisioned_keys",
"_gitconfig_validate_value",
"_provision_dynamic_key",
+50 -4
View File
@@ -9,10 +9,17 @@ imported (`deploy_key_provisioner`) to keep its cost off the host path.
from __future__ import annotations
import os
import dataclasses
from pathlib import Path
from typing import TYPE_CHECKING
from .errors import MissingEnvVarError
from .log import info
from .manifest import ManifestBottle, ManifestGitEntry
from .git_gate_render import GitGateUpstream
if TYPE_CHECKING:
from .git_gate import GitGatePlan
def _provision_dynamic_key(
entry: ManifestGitEntry,
@@ -28,9 +35,10 @@ def _provision_dynamic_key(
pk = entry.Key
token = os.environ.get(pk.forge_token_env)
if token is None:
raise RuntimeError(
raise MissingEnvVarError(
pk.forge_token_env,
f"git-gate.repos[{entry.Name!r}] key.forge_token_env"
f" = {pk.forge_token_env!r}: env var is not set"
f" = {pk.forge_token_env!r}: env var is not set",
)
api_url = pk.api_url or f"https://{entry.UpstreamHost}"
provisioner = get_provisioner(pk.provider, token, api_url)
@@ -72,10 +80,11 @@ def revoke_git_gate_provisioned_keys(bottle: ManifestBottle, stage_dir: Path) ->
key_id = id_file.read_text().strip()
token = os.environ.get(pk.forge_token_env)
if token is None:
raise RuntimeError(
raise MissingEnvVarError(
pk.forge_token_env,
f"git-gate.repos[{entry.Name!r}] key.forge_token_env"
f" = {pk.forge_token_env!r}: env var is not set;"
f" cannot revoke deploy key {key_id}"
f" cannot revoke deploy key {key_id}",
)
api_url = pk.api_url or f"https://{entry.UpstreamHost}"
provisioner = get_provisioner(pk.provider, token, api_url)
@@ -95,8 +104,45 @@ def _resolve_identity_file(entry: ManifestGitEntry, slug: str, stage_dir: Path)
return entry.IdentityFile
def provision_git_gate_dynamic_keys(
bottle: ManifestBottle,
plan: "GitGatePlan",
stage_dir: Path,
) -> "GitGatePlan":
"""Provision dynamic git-gate keys and return an updated plan.
This runs during backend launch, after the operator confirms the
preflight. Plan preparation intentionally stays side-effect-light:
dry-runs and aborted launches must not create remote deploy keys.
"""
if not plan.upstreams:
return plan
upstreams_by_name: dict[str, GitGateUpstream] = {
upstream.name: upstream for upstream in plan.upstreams
}
updated: list[GitGateUpstream] = []
for entry in bottle.git:
upstream = upstreams_by_name.get(entry.Name)
if upstream is None:
continue
if entry.Key.provider == "gitea":
identity_file = _provision_dynamic_key(entry, plan.slug, stage_dir)
upstream = dataclasses.replace(upstream, identity_file=identity_file)
updated.append(upstream)
if len(updated) != len(plan.upstreams):
updated_names = {u.name for u in updated}
for upstream in plan.upstreams:
if upstream.name not in updated_names:
updated.append(upstream)
return dataclasses.replace(plan, upstreams=tuple(updated))
__all__ = [
"revoke_git_gate_provisioned_keys",
"provision_git_gate_dynamic_keys",
"_provision_dynamic_key",
"_resolve_identity_file",
]
+48 -44
View File
@@ -228,19 +228,22 @@ supervise_gitleaks_allow() {
fi
proposal_id=$(
GITLEAKS_ALLOW_REF="$ref" python3 - "$report_file" <<'PY'
PYTHONPATH="/app${PYTHONPATH:+:$PYTHONPATH}" GITLEAKS_ALLOW_REF="$ref" python3 - "$report_file" <<'PY'
import datetime
import hashlib
import json
import os
import sys
import uuid
from pathlib import Path
try:
import supervise as _sv
except ImportError:
from bot_bottle import supervise as _sv
report_path = Path(sys.argv[1])
queue_dir = os.environ.get("SUPERVISE_QUEUE_DIR", "")
slug = os.environ.get("SUPERVISE_BOTTLE_SLUG", "")
if not queue_dir or not slug:
if not slug:
sys.exit(2)
try:
@@ -277,31 +280,19 @@ for i, finding in enumerate(raw, 1):
])
payload = "\n".join(lines).rstrip() + "\n"
proposal_id = str(uuid.uuid4())
proposal = {
"id": proposal_id,
"bottle_slug": slug,
"tool": "gitleaks-allow",
"proposed_file": payload,
"justification": (
proposal = _sv.Proposal.new(
bottle_slug=slug,
tool=_sv.TOOL_GITLEAKS_ALLOW,
proposed_file=payload,
justification=(
"git-gate found gitleaks findings hidden by # gitleaks:allow; "
"approve only for dummy test fixtures or confirmed false positives"
),
"arrival_timestamp": datetime.datetime.now(
datetime.timezone.utc
).isoformat(),
"current_file_hash": hashlib.sha256(payload.encode("utf-8")).hexdigest(),
}
queue = Path(queue_dir)
queue.mkdir(parents=True, exist_ok=True)
path = queue / f"{proposal_id}.proposal.json"
tmp = path.with_suffix(path.suffix + ".tmp")
with tmp.open("w", encoding="utf-8") as f:
json.dump(proposal, f, indent=2)
f.write("\n")
os.chmod(tmp, 0o600)
os.replace(tmp, path)
print(proposal_id)
current_file_hash=hashlib.sha256(payload.encode("utf-8")).hexdigest(),
now=datetime.datetime.now(datetime.timezone.utc),
)
_sv.write_proposal(proposal)
print(proposal.id)
PY
)
rc=$?
@@ -314,8 +305,7 @@ PY
return 1
fi
queue_dir=${SUPERVISE_QUEUE_DIR:-}
response_file="$queue_dir/${proposal_id}.response.json"
slug=${SUPERVISE_BOTTLE_SLUG:-}
timeout=${SUPERVISE_GITLEAKS_ALLOW_TIMEOUT_SECONDS:-300}
case "$timeout" in
''|*[!0-9]*)
@@ -327,26 +317,41 @@ PY
echo "git-gate: approve with './cli.py supervise' to continue this push" >&2
waited=0
while [ "$waited" -lt "$timeout" ]; do
if [ -f "$response_file" ]; then
status=$(python3 - "$response_file" <<'PY'
import json
status=$(PYTHONPATH="/app${PYTHONPATH:+:$PYTHONPATH}" python3 - "$slug" "$proposal_id" <<'PY'
import sys
try:
with open(sys.argv[1], encoding="utf-8") as f:
raw = json.load(f)
except (OSError, json.JSONDecodeError):
sys.exit(1)
status = raw.get("status")
if not isinstance(status, str):
sys.exit(1)
print(status)
import supervise as _sv
except ImportError:
from bot_bottle import supervise as _sv
slug = sys.argv[1]
try:
response = _sv.read_response(slug, sys.argv[2])
except FileNotFoundError:
sys.exit(2)
print(response.status)
PY
) || status=""
)
rc=$?
if [ "$rc" -eq 2 ]; then
status=""
elif [ "$rc" -ne 0 ]; then
status="invalid"
fi
if [ -n "$status" ]; then
case "$status" in
approved|modified)
mkdir -p "$queue_dir/processed"
mv -f "$queue_dir/${proposal_id}.proposal.json" "$queue_dir/processed/" 2>/dev/null || true
mv -f "$queue_dir/${proposal_id}.response.json" "$queue_dir/processed/" 2>/dev/null || true
PYTHONPATH="/app${PYTHONPATH:+:$PYTHONPATH}" python3 - "$slug" "$proposal_id" <<'PY' || true
import sys
try:
import supervise as _sv
except ImportError:
from bot_bottle import supervise as _sv
_sv.archive_proposal(sys.argv[1], sys.argv[2])
PY
echo "git-gate: supervisor approved # gitleaks:allow for $ref" >&2
return 0
;;
@@ -499,4 +504,3 @@ if ! git -C "$repo_dir" rev-parse --verify HEAD >/dev/null 2>&1; then
fi
exit 0
"""
+7 -2
View File
@@ -16,11 +16,16 @@ from http.server import BaseHTTPRequestHandler, ThreadingHTTPServer
from pathlib import Path
from urllib.parse import urlsplit
from .git_gate import GIT_GATE_TIMEOUT_SECS
DEFAULT_PORT = 9420
# Mirrors git_gate_render.GIT_GATE_TIMEOUT_SECS. Duplicated rather than
# imported: this module ships as a flat top-level sibling in the sidecar
# bundle image (see Dockerfile.sidecars), not as part of the bot_bottle
# package, so `bot_bottle.git_gate` and its dependency chain aren't
# available at runtime.
GIT_GATE_TIMEOUT_SECS = 15
# Bound memory use while still allowing ordinary git push packfiles.
MAX_BODY_BYTES = 100 * 1024 * 1024
+12 -122
View File
@@ -62,15 +62,25 @@ from dataclasses import dataclass, field, replace
from pathlib import Path
from typing import Mapping
from .log import warn
from .manifest_util import ManifestError, as_json_object
from .manifest_agent import ManifestAgent, ManifestAgentProvider
from .manifest_bottle import ManifestBottle
from .manifest_egress import (
EGRESS_AUTH_SCHEMES,
ManifestEgressConfig,
ManifestEgressRoute,
)
from .manifest_git import ManifestGitEntry, ManifestGitUser, ManifestKeyConfig, parse_git_gate_config
from .manifest_schema import BOTTLE_KEYS
from .manifest_extends import merge_bottles_runtime, resolve_bottles
from .manifest_git import ManifestGitEntry, ManifestGitUser, ManifestKeyConfig
from .manifest_loader import (
check_stale_json,
load_bottle_chain_from_dir,
scan_agent_names,
scan_bottle_names,
)
from .manifest_schema import validate_agent_frontmatter_keys
from .yaml_subset import YamlSubsetError, parse_frontmatter
# Re-export everything that callers currently import from this module.
__all__ = [
@@ -89,10 +99,6 @@ __all__ = [
]
def _empty_str_dict() -> dict[str, str]:
return {}
def _section_dict(value: object, label: str) -> dict[str, object]:
"""Like as_json_object but treats absent/null as an empty section."""
if value is None:
@@ -100,107 +106,6 @@ def _section_dict(value: object, label: str) -> dict[str, object]:
return as_json_object(value, label)
@dataclass(frozen=True)
class ManifestBottle:
env: Mapping[str, str] = field(default_factory=_empty_str_dict)
agent_provider: ManifestAgentProvider = field(default_factory=ManifestAgentProvider)
git: tuple[ManifestGitEntry, ...] = ()
# Per-bottle git identity (issue #86). Empty default — bottles
# that don't set `git-gate.user:` in the manifest skip the
# `git config --global` step entirely. A bottle can declare a user
# identity without any git-gate.repos upstreams, and vice versa.
git_user: ManifestGitUser = field(default_factory=ManifestGitUser)
egress: ManifestEgressConfig = field(default_factory=ManifestEgressConfig)
# Per-bottle stuck-recovery sidecar (PRD 0013). When true (the
# default, issue #249), the launch step brings up a supervise
# sidecar that exposes egress MCP tools to the agent. Set
# `supervise: false` to skip the sidecar.
supervise: bool = True
@classmethod
def from_dict(cls, name: str, raw: object) -> "ManifestBottle":
d = as_json_object(raw, f"bottle '{name}'")
if "runtime" in d:
raise ManifestError(
f"bottle '{name}' has a 'runtime' field, which is no longer "
f"supported. gVisor (runsc) is now auto-detected by the "
f"backend; remove the 'runtime' field from the bottle "
f"definition."
)
if "ssh" in d:
raise ManifestError(
f"bottle '{name}' has an 'ssh' field, which has been removed "
f"(PRD 0009). Declare upstreams under 'git-gate.repos' with "
f"url + identity + host_key; the git-gate sidecar (PRD 0008) "
f"holds the credential and gitleaks-scans pushes."
)
if "git" in d:
raise ManifestError(
f"bottle '{name}' uses 'git' which has been replaced by "
f"'git-gate' (PRD 0047). Move git.user → git-gate.user "
f"and git.remotes → git-gate.repos (fields: url, identity, host_key)."
)
if "git_user" in d:
raise ManifestError(
f"bottle '{name}' has a 'git_user' field, which has been "
f"removed. Move it under 'git-gate.user'."
)
unknown = set(d.keys()) - BOTTLE_KEYS
if unknown:
allowed = ", ".join(sorted(BOTTLE_KEYS))
raise ManifestError(
f"bottle '{name}' has unknown key(s) {sorted(unknown)}; "
f"allowed keys are {allowed}."
)
env: dict[str, str] = {}
env_raw = d.get("env")
if env_raw is not None:
env_dict = as_json_object(env_raw, f"bottle '{name}' env")
for var, value in env_dict.items():
if not isinstance(value, str):
raise ManifestError(
f"env entry {var} in bottle '{name}' must be a JSON string "
f"(was {type(value).__name__}). Use \"?<message>\" for prompt-at-runtime."
)
env[var] = value
git: tuple[ManifestGitEntry, ...] = ()
git_user = ManifestGitUser()
git_raw = d.get("git-gate")
if git_raw is not None:
git, git_user = parse_git_gate_config(name, git_raw)
agent_provider = (
ManifestAgentProvider.from_dict(name, d["agent_provider"])
if "agent_provider" in d
else ManifestAgentProvider()
)
egress = (
ManifestEgressConfig.from_dict(name, d["egress"])
if "egress" in d
else ManifestEgressConfig()
)
supervise_raw = d.get("supervise", True)
if not isinstance(supervise_raw, bool):
raise ManifestError(
f"bottle '{name}' supervise must be a boolean "
f"(was {type(supervise_raw).__name__})"
)
return cls(
env=env, agent_provider=agent_provider, git=git,
git_user=git_user, egress=egress, supervise=supervise_raw,
)
def _merge_git_user(
agent_user: ManifestGitUser, base_user: ManifestGitUser
) -> ManifestGitUser:
@@ -237,8 +142,6 @@ def _resolve_effective_bottle_eager(
When bottle_names is non-empty they are merged in order. When empty, falls
back to agent.bottle. Raises ManifestError when neither is set."""
from .manifest_extends import merge_bottles_runtime
if bottle_names:
resolved: list[ManifestBottle] = []
for bn in bottle_names:
@@ -270,9 +173,6 @@ def _resolve_effective_bottle_lazy(
When bottle_names is non-empty they are resolved from disk and merged in
order. When empty, falls back to agent_bottle. Raises ManifestError when
neither is set."""
from .manifest_extends import merge_bottles_runtime
from .manifest_loader import load_bottle_chain_from_dir
if bottle_names:
resolved = [load_bottle_chain_from_dir(bn, bottles_dir) for bn in bottle_names]
return merge_bottles_runtime(resolved)
@@ -358,8 +258,6 @@ class ManifestIndex:
home_md = home_dir / ".bot-bottle"
cwd_md = cwd_dir / ".bot-bottle"
from .manifest_loader import check_stale_json
check_stale_json(home_dir, home_md, "$HOME")
if cwd_dir.resolve() != home_dir.resolve():
check_stale_json(cwd_dir, cwd_md, "$CWD")
@@ -399,7 +297,6 @@ class ManifestIndex:
files = sorted(stale_bottles.glob("*.md"))
if files:
names = ", ".join(p.name for p in files)
from .log import warn
warn(
f"ignoring bottle file(s) under "
f"{stale_bottles}: {names}. Bottles can only "
@@ -421,7 +318,6 @@ class ManifestIndex:
raw_bottles: dict[str, dict[str, object]] = {}
for n, b in raw_bottles_obj.items():
raw_bottles[n] = as_json_object(b, f"bottle '{n}'")
from .manifest_extends import resolve_bottles
bottles = resolve_bottles(raw_bottles)
@@ -439,7 +335,6 @@ class ManifestIndex:
filenames without reading their content. In eager mode (from
from_json_obj) it returns the pre-parsed bottles' names."""
if self.home_md is not None:
from .manifest_loader import scan_bottle_names
return scan_bottle_names(self.home_md / "bottles")
return sorted(self.bottles.keys())
@@ -451,7 +346,6 @@ class ManifestIndex:
filenames without reading their content. In eager mode (from
from_json_obj) it returns the pre-parsed agents' names."""
if self.home_md is not None:
from .manifest_loader import scan_agent_names
home_names = set(scan_agent_names(self.home_md / "agents").keys())
cwd_names: set[str] = set()
if self.cwd_md is not None:
@@ -509,10 +403,6 @@ class ManifestIndex:
"""Lazy path (resolve/from_md_dirs): read and parse the agent file and
its bottle chain from disk for the first time here."""
assert self.home_md is not None # guaranteed by load_for_agent dispatch
from .manifest_loader import scan_agent_names
from .manifest_schema import validate_agent_frontmatter_keys
from .yaml_subset import YamlSubsetError, parse_frontmatter
# Locate the agent file; cwd wins over home on name collision.
home_agents = scan_agent_names(self.home_md / "agents")
cwd_agents: dict[str, Path] = {}
+11 -1
View File
@@ -8,7 +8,7 @@ from typing import cast
from .agent_provider import PROVIDER_TEMPLATES
from .manifest_util import ManifestError, as_json_object
from .manifest_git import ManifestGitUser
from .manifest_schema import AGENT_MODEL_KEYS
from .manifest_schema import AGENT_MODEL_KEYS, is_valid_entity_name
@dataclass(frozen=True)
@@ -161,6 +161,16 @@ class ManifestAgent:
f"agent '{name}' skills[{i}] must be a string "
f"(was {type(skill).__name__})"
)
# Skill names become host/guest path segments and are
# interpolated into provisioning shell commands, so they
# must fit the same kebab-case convention as bottle/agent
# filenames — rejecting anything that could break out of a
# path segment or inject shell metacharacters.
if not is_valid_entity_name(skill):
raise ManifestError(
f"agent '{name}' skills[{i}] {skill!r} is not a valid "
f"skill name; must match [a-z][a-z0-9-]*"
)
collected.append(skill)
skills = tuple(collected)
+129
View File
@@ -0,0 +1,129 @@
"""The `ManifestBottle` value type.
Split out of `manifest.py` so the `extends:`/loader resolvers can import it
without a circular dependency: `manifest.py` imports those resolvers, while
they only need this value type. Everything here depends on leaf modules
(`manifest_util`, `manifest_agent`, `manifest_egress`, `manifest_git`,
`manifest_schema`), so this module sits at the bottom of the manifest layer.
`manifest.py` re-exports `ManifestBottle`, so existing
`from .manifest import ManifestBottle` callers are unaffected.
"""
from __future__ import annotations
from dataclasses import dataclass, field
from typing import Mapping
from .manifest_util import ManifestError, as_json_object
from .manifest_agent import ManifestAgentProvider
from .manifest_egress import ManifestEgressConfig
from .manifest_git import ManifestGitEntry, ManifestGitUser, parse_git_gate_config
from .manifest_schema import BOTTLE_KEYS
__all__ = ["ManifestBottle"]
def _empty_str_dict() -> dict[str, str]:
return {}
@dataclass(frozen=True)
class ManifestBottle:
env: Mapping[str, str] = field(default_factory=_empty_str_dict)
agent_provider: ManifestAgentProvider = field(default_factory=ManifestAgentProvider)
git: tuple[ManifestGitEntry, ...] = ()
# Per-bottle git identity (issue #86). Empty default — bottles
# that don't set `git-gate.user:` in the manifest skip the
# `git config --global` step entirely. A bottle can declare a user
# identity without any git-gate.repos upstreams, and vice versa.
git_user: ManifestGitUser = field(default_factory=ManifestGitUser)
egress: ManifestEgressConfig = field(default_factory=ManifestEgressConfig)
# Per-bottle stuck-recovery sidecar (PRD 0013). When true (the
# default, issue #249), the launch step brings up a supervise
# sidecar that exposes egress MCP tools to the agent. Set
# `supervise: false` to skip the sidecar.
supervise: bool = True
@classmethod
def from_dict(cls, name: str, raw: object) -> "ManifestBottle":
d = as_json_object(raw, f"bottle '{name}'")
if "runtime" in d:
raise ManifestError(
f"bottle '{name}' has a 'runtime' field, which is no longer "
f"supported. gVisor (runsc) is now auto-detected by the "
f"backend; remove the 'runtime' field from the bottle "
f"definition."
)
if "ssh" in d:
raise ManifestError(
f"bottle '{name}' has an 'ssh' field, which has been removed "
f"(PRD 0009). Declare upstreams under 'git-gate.repos' with "
f"url + identity + host_key; the git-gate sidecar (PRD 0008) "
f"holds the credential and gitleaks-scans pushes."
)
if "git" in d:
raise ManifestError(
f"bottle '{name}' uses 'git' which has been replaced by "
f"'git-gate' (PRD 0047). Move git.user → git-gate.user "
f"and git.remotes → git-gate.repos (fields: url, identity, host_key)."
)
if "git_user" in d:
raise ManifestError(
f"bottle '{name}' has a 'git_user' field, which has been "
f"removed. Move it under 'git-gate.user'."
)
unknown = set(d.keys()) - BOTTLE_KEYS
if unknown:
allowed = ", ".join(sorted(BOTTLE_KEYS))
raise ManifestError(
f"bottle '{name}' has unknown key(s) {sorted(unknown)}; "
f"allowed keys are {allowed}."
)
env: dict[str, str] = {}
env_raw = d.get("env")
if env_raw is not None:
env_dict = as_json_object(env_raw, f"bottle '{name}' env")
for var, value in env_dict.items():
if not isinstance(value, str):
raise ManifestError(
f"env entry {var} in bottle '{name}' must be a JSON string "
f"(was {type(value).__name__}). Use \"?<message>\" for prompt-at-runtime."
)
env[var] = value
git: tuple[ManifestGitEntry, ...] = ()
git_user = ManifestGitUser()
git_raw = d.get("git-gate")
if git_raw is not None:
git, git_user = parse_git_gate_config(name, git_raw)
agent_provider = (
ManifestAgentProvider.from_dict(name, d["agent_provider"])
if "agent_provider" in d
else ManifestAgentProvider()
)
egress = (
ManifestEgressConfig.from_dict(name, d["egress"])
if "egress" in d
else ManifestEgressConfig()
)
supervise_raw = d.get("supervise", True)
if not isinstance(supervise_raw, bool):
raise ManifestError(
f"bottle '{name}' supervise must be a boolean "
f"(was {type(supervise_raw).__name__})"
)
return cls(
env=env, agent_provider=agent_provider, git=git,
git_user=git_user, egress=egress, supervise=supervise_raw,
)
+4 -28
View File
@@ -2,11 +2,10 @@
from __future__ import annotations
from typing import TYPE_CHECKING
if TYPE_CHECKING:
from .manifest import ManifestBottle
from .manifest_egress import ManifestEgressConfig
from .manifest_bottle import ManifestBottle
from .manifest_egress import ManifestEgressConfig, validate_egress_routes
from .manifest_git import ManifestGitUser, parse_git_gate_config
from .manifest_util import ManifestError, as_json_object
def merge_bottles_runtime(bottles: "list[ManifestBottle]") -> "ManifestBottle":
@@ -27,9 +26,6 @@ def merge_bottles_runtime(bottles: "list[ManifestBottle]") -> "ManifestBottle":
def _merge_two_bottles_runtime(base: "ManifestBottle", override: "ManifestBottle") -> "ManifestBottle":
from .manifest import ManifestBottle, ManifestGitUser
from .manifest_egress import ManifestEgressConfig
merged_env = {**base.env, **override.env}
merged_git_user = ManifestGitUser(
@@ -81,8 +77,6 @@ def _resolve_one_bottle(
repos_cache: dict[str, dict[str, object]],
seen: tuple[str, ...],
) -> ManifestBottle:
from .manifest import ManifestBottle, ManifestError
if name in cache:
return cache[name]
if name in seen:
@@ -174,11 +168,6 @@ def _fold_two_bottles(
later_repos_raw: dict[str, object],
) -> tuple[ManifestBottle, dict[str, object]]:
"""Combine two resolved parent bottles; later wins over earlier."""
from .manifest import ManifestBottle, ManifestGitUser
from .manifest_egress import ManifestEgressConfig
from .manifest_git import parse_git_gate_config
from .manifest_util import as_json_object
merged_env = {**earlier.env, **later.env}
merged_git_user = ManifestGitUser(
@@ -227,10 +216,6 @@ def _merge_bottles(
name: str,
) -> ManifestBottle:
"""Apply PRD 0025 merge rules."""
from .manifest import ManifestBottle, ManifestGitUser
from .manifest_egress import validate_egress_routes
from .manifest_util import as_json_object
# git-gate.repos: when the child declares repos, inject the already
# name-merged repo set (computed by _resolve_repos_raw) so the child
# parses with the full inherited+overridden list (issue #237).
@@ -303,8 +288,6 @@ def _resolve_repos_raw(
inherits the parent's set verbatim; an explicit empty dict clears it.
Otherwise parent and child unite by name, with same-name entries
field-merged (parent fields are defaults, child fields win)."""
from .manifest_util import as_json_object
if not _child_declares_git_gate_repos(child_raw):
return parent_repos
child_repos = _declared_repos_raw(child_raw)
@@ -324,8 +307,6 @@ def _resolve_repos_raw(
def _declared_repos_raw(child_raw: dict[str, object]) -> dict[str, object]:
"""Return the child's explicitly declared git-gate.repos as raw dicts,
or an empty dict when none are declared."""
from .manifest_util import as_json_object
if not _child_declares_git_gate_repos(child_raw):
return {}
git_raw = as_json_object(child_raw.get("git-gate", {}), "child git-gate")
@@ -333,8 +314,6 @@ def _declared_repos_raw(child_raw: dict[str, object]) -> dict[str, object]:
def _child_declares_git_gate_repos(child_raw: dict[str, object]) -> bool:
from .manifest_util import as_json_object
git_raw = child_raw.get("git-gate")
if git_raw is None:
return False
@@ -347,9 +326,6 @@ def _merge_egress(
child: ManifestEgressConfig,
child_raw: dict[str, object],
) -> ManifestEgressConfig:
from .manifest_egress import ManifestEgressConfig
from .manifest_util import as_json_object
child_egress_raw = as_json_object(child_raw.get("egress"), "child egress")
routes = parent.routes + child.routes
log = child.Log if "log" in child_egress_raw else parent.Log
+2 -6
View File
@@ -3,9 +3,10 @@
from __future__ import annotations
from pathlib import Path
from typing import TYPE_CHECKING
from .log import warn
from .manifest_bottle import ManifestBottle
from .manifest_extends import resolve_bottles
from .manifest_schema import (
entity_name_from_path,
validate_bottle_frontmatter_keys,
@@ -13,9 +14,6 @@ from .manifest_schema import (
from .manifest_util import ManifestError
from .yaml_subset import YamlSubsetError, parse_frontmatter
if TYPE_CHECKING:
from .manifest import ManifestBottle
def check_stale_json(dir_path: Path, md_dir: Path, label: str) -> None:
"""Die if `<dir_path>/bot-bottle.json` exists but `md_dir` does
@@ -78,8 +76,6 @@ def load_bottle_chain_from_dir(
Only the files in the extends chain are read unrelated bottle files
are never touched. Raises ManifestError on parse or validation failure."""
from .manifest_extends import resolve_bottles
raws: dict[str, dict[str, object]] = {}
to_load = [bottle_name]
while to_load:
+8 -1
View File
@@ -33,13 +33,20 @@ AGENT_KEYS = (
AGENT_MODEL_KEYS = AGENT_KEYS | frozenset({"prompt"})
def is_valid_entity_name(name: str) -> bool:
"""True if `name` fits the kebab-case `[a-z][a-z0-9-]*` convention
shared by bottle/agent filenames and skill names. Names that satisfy
this are also safe to interpolate into a host/guest path segment."""
return bool(_FILENAME_RX.match(name))
def entity_name_from_path(path: Path) -> str | None:
"""Return the entity name implied by the filename, or None if the
filename does not fit the [a-z][a-z0-9-]* convention."""
if path.suffix != ".md":
return None
stem = path.stem
if not _FILENAME_RX.match(stem):
if not is_valid_entity_name(stem):
return None
return stem
+37
View File
@@ -0,0 +1,37 @@
"""SQLite migration runner for bot-bottle stores."""
from __future__ import annotations
import sqlite3
class TableMigrations:
"""Runs a sequential list of DDL migrations tracked by schema_key in schema_versions."""
def __init__(self, schema_key: str, migrations: list[str]) -> None:
self.schema_key = schema_key
self.migrations = migrations
def apply(self, conn: sqlite3.Connection) -> None:
conn.execute(
"""
CREATE TABLE IF NOT EXISTS schema_versions (
module TEXT PRIMARY KEY,
version INTEGER NOT NULL DEFAULT 0
)
"""
)
row = conn.execute(
"SELECT version FROM schema_versions WHERE module = ?",
(self.schema_key,),
).fetchone()
version = row[0] if row else 0
for i, sql in enumerate(self.migrations[version:], start=version + 1):
conn.execute(sql)
conn.execute(
"INSERT OR REPLACE INTO schema_versions (module, version) VALUES (?, ?)",
(self.schema_key, i),
)
__all__ = ["TableMigrations"]
+215
View File
@@ -0,0 +1,215 @@
"""SQLite-backed queue store for supervise proposals and responses (PRD 0013)."""
from __future__ import annotations
import os
import sqlite3
from pathlib import Path
try:
from .supervise_types import Proposal, Response, host_db_path
from .db_store import DbStore
from .migrations import TableMigrations
except ImportError:
from supervise_types import Proposal, Response, host_db_path # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
from db_store import DbStore # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
from migrations import TableMigrations # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
class QueueStore(DbStore):
"""SQLite-backed persistent store for supervise proposals and responses."""
def __init__(self, queue_key: str, db_path: Path | None = None) -> None:
self.queue_key = queue_key
if db_path is not None:
resolved = db_path
else:
# In the sidecar container SUPERVISE_DB_PATH points at the
# bind-mounted host DB. On the host this env var is never set,
# so we always fall through to host_db_path().
env_path = os.environ.get("SUPERVISE_DB_PATH", "").strip()
resolved = Path(env_path) if env_path else host_db_path()
# One entry per schema version: migrations[0] brings a fresh DB to
# version 1, [1] to version 2, etc. Add new entries at the end; never
# edit existing ones.
migrations = TableMigrations("queue_store", [
# v1 — proposals table
"""
CREATE TABLE IF NOT EXISTS supervise_proposals (
queue_key TEXT NOT NULL,
id TEXT NOT NULL,
bottle_slug TEXT NOT NULL,
tool TEXT NOT NULL,
proposed_file TEXT NOT NULL,
justification TEXT NOT NULL,
arrival_timestamp TEXT NOT NULL,
current_file_hash TEXT NOT NULL,
archived INTEGER NOT NULL DEFAULT 0,
PRIMARY KEY (queue_key, id)
)
""",
# v2 — responses table
"""
CREATE TABLE IF NOT EXISTS supervise_responses (
queue_key TEXT NOT NULL,
proposal_id TEXT NOT NULL,
status TEXT NOT NULL,
notes TEXT NOT NULL,
final_file TEXT,
archived INTEGER NOT NULL DEFAULT 0,
PRIMARY KEY (queue_key, proposal_id)
)
""",
])
super().__init__(resolved, migrations)
def write_proposal(self, proposal: Proposal) -> Path:
with self._connect() as conn:
conn.execute(
"""
INSERT OR REPLACE INTO supervise_proposals (
queue_key, id, bottle_slug, tool, proposed_file, justification,
arrival_timestamp, current_file_hash, archived
) VALUES (?, ?, ?, ?, ?, ?, ?, ?, 0)
""",
(
self.queue_key,
proposal.id,
proposal.bottle_slug,
proposal.tool,
proposal.proposed_file,
proposal.justification,
proposal.arrival_timestamp,
proposal.current_file_hash,
),
)
self._chmod()
return self.db_path
def read_proposal(self, proposal_id: str) -> Proposal:
with self._connect() as conn:
row = conn.execute(
"""
SELECT * FROM supervise_proposals
WHERE queue_key = ? AND id = ? AND archived = 0
""",
(self.queue_key, proposal_id),
).fetchone()
if row is None:
raise FileNotFoundError(proposal_id)
return self._row_to_proposal(row)
def list_pending_proposals(self) -> list[Proposal]:
if not self.db_path.is_file():
return []
with self._connect() as conn:
rows = conn.execute(
"""
SELECT p.* FROM supervise_proposals p
WHERE p.archived = 0
AND p.queue_key = ?
AND NOT EXISTS (
SELECT 1 FROM supervise_responses r
WHERE r.queue_key = p.queue_key
AND r.proposal_id = p.id
AND r.archived = 0
)
ORDER BY p.arrival_timestamp, p.id
""",
(self.queue_key,),
).fetchall()
return [self._row_to_proposal(row) for row in rows]
def list_all_pending_proposals(self) -> list[Proposal]:
if not self.db_path.is_file():
return []
with self._connect() as conn:
rows = conn.execute(
"""
SELECT p.* FROM supervise_proposals p
WHERE p.archived = 0
AND NOT EXISTS (
SELECT 1 FROM supervise_responses r
WHERE r.queue_key = p.queue_key
AND r.proposal_id = p.id
AND r.archived = 0
)
ORDER BY p.arrival_timestamp, p.id
"""
).fetchall()
return [self._row_to_proposal(row) for row in rows]
def write_response(self, response: Response) -> Path:
with self._connect() as conn:
conn.execute(
"""
INSERT OR REPLACE INTO supervise_responses (
queue_key, proposal_id, status, notes, final_file, archived
) VALUES (?, ?, ?, ?, ?, 0)
""",
(
self.queue_key,
response.proposal_id,
response.status,
response.notes,
response.final_file,
),
)
self._chmod()
return self.db_path
def read_response(self, proposal_id: str) -> Response:
with self._connect() as conn:
row = conn.execute(
"""
SELECT * FROM supervise_responses
WHERE queue_key = ? AND proposal_id = ? AND archived = 0
""",
(self.queue_key, proposal_id),
).fetchone()
if row is None:
raise FileNotFoundError(proposal_id)
return self._row_to_response(row)
def archive_proposal(self, proposal_id: str) -> None:
if not self.db_path.is_file():
return
with self._connect() as conn:
conn.execute(
"""
UPDATE supervise_proposals SET archived = 1
WHERE queue_key = ? AND id = ?
""",
(self.queue_key, proposal_id),
)
conn.execute(
"""
UPDATE supervise_responses SET archived = 1
WHERE queue_key = ? AND proposal_id = ?
""",
(self.queue_key, proposal_id),
)
@staticmethod
def _row_to_proposal(row: sqlite3.Row) -> Proposal:
return Proposal(
id=row["id"],
bottle_slug=row["bottle_slug"],
tool=row["tool"],
proposed_file=row["proposed_file"],
justification=row["justification"],
arrival_timestamp=row["arrival_timestamp"],
current_file_hash=row["current_file_hash"],
)
@staticmethod
def _row_to_response(row: sqlite3.Row) -> Response:
return Response(
proposal_id=row["proposal_id"],
status=row["status"],
notes=row["notes"],
final_file=row["final_file"],
)
__all__ = ["QueueStore"]
+60
View File
@@ -0,0 +1,60 @@
"""Singleton manager for all bot-bottle SQLite stores (PRD 0013)."""
from __future__ import annotations
from pathlib import Path
try:
from .audit_store import AuditStore
from .queue_store import QueueStore
except ImportError:
from audit_store import AuditStore # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
from queue_store import QueueStore # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
_instance: StoreManager | None = None
class StoreManager:
"""Owns db_path and delegates migrate/is_migrated across all stores.
Use instance() for normal access. Call reset(db_path) in tests to swap
the singleton to a temp path, then reset() with no args to restore the
default."""
def __init__(self, db_path: Path | None = None) -> None:
if db_path is None:
# Lazy import to avoid a circular dependency: supervise imports
# StoreManager at module level; StoreManager must not import
# supervise at module level in return.
try:
from .supervise import host_db_path
except ImportError:
from supervise import host_db_path # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
db_path = host_db_path()
self.db_path = db_path
@classmethod
def instance(cls) -> StoreManager:
global _instance
if _instance is None:
_instance = cls()
return _instance
@classmethod
def reset(cls, db_path: Path | None = None) -> None:
"""Replace the singleton. Pass db_path for test isolation; omit to restore default."""
global _instance
_instance = cls(db_path)
def is_migrated(self) -> bool:
return (
QueueStore("", self.db_path).is_migrated()
and AuditStore(self.db_path).is_migrated()
)
def migrate(self) -> None:
QueueStore("", self.db_path).migrate()
AuditStore(self.db_path).migrate()
__all__ = ["StoreManager"]
+112 -349
View File
@@ -9,15 +9,14 @@ calls when it needs an operator-reviewed egress change:
Each tool call: the agent passes the full proposed file plus a
justification text. The sidecar validates the proposal syntactically,
writes it to the host's per-bottle queue dir, and holds the tool-call
writes it to the host SQLite queue table, and holds the tool-call
connection open. The operator's supervise TUI
(bot_bottle.cli.supervise) sees the proposal, accepts
approve / modify / reject, and writes a response file alongside the
proposal. The sidecar sees the response and returns `{status, notes}`
to the agent.
approve / modify / reject, and writes a response row. The sidecar sees
the response and returns `{status, notes}` to the agent.
This module defines the host-side library: dataclasses for the queue
file shapes, queue read/write helpers, the audit log writer, and the
record shapes, queue read/write helpers, the audit log writer, and the
diff renderer. The in-container sidecar lives in
bot_bottle/supervise_server.py; the supervise daemon's container
lifecycle is owned by the sidecar bundle (PRD 0024).
@@ -31,37 +30,56 @@ remediation engines that wire real config changes land in PRDs 0014,
from __future__ import annotations
import dataclasses
import difflib
import hashlib
import json
import os
import time
import uuid
from abc import ABC
from dataclasses import dataclass
from datetime import datetime, timezone
from pathlib import Path
try:
from .supervise_types import (
ACTION_OPERATOR_EDIT,
AuditEntry,
HOST_DB_FILENAME,
Proposal,
Response,
STATUSES,
STATUS_APPROVED,
STATUS_MODIFIED,
STATUS_REJECTED,
TOOLS,
TOOL_EGRESS_ALLOW,
TOOL_EGRESS_BLOCK,
TOOL_EGRESS_TOKEN_ALLOW,
TOOL_GITLEAKS_ALLOW,
TOOL_LIST_EGRESS_ROUTES,
bot_bottle_root,
)
except ImportError:
from supervise_types import ( # type: ignore[import-not-found,no-redef] # pylint: disable=import-error,no-name-in-module
ACTION_OPERATOR_EDIT,
AuditEntry,
HOST_DB_FILENAME,
Proposal,
Response,
STATUSES,
STATUS_APPROVED,
STATUS_MODIFIED,
STATUS_REJECTED,
TOOLS,
TOOL_EGRESS_ALLOW,
TOOL_EGRESS_BLOCK,
TOOL_EGRESS_TOKEN_ALLOW,
TOOL_GITLEAKS_ALLOW,
TOOL_LIST_EGRESS_ROUTES,
bot_bottle_root,
)
SUPERVISE_HOSTNAME = "supervise"
SUPERVISE_PORT = 9100
TOOL_EGRESS_BLOCK = "egress-block"
TOOL_EGRESS_ALLOW = "egress-allow"
TOOL_GITLEAKS_ALLOW = "gitleaks-allow"
# Written directly by the egress addon (not an agent-facing MCP tool) when an
# outbound DLP token block is routed to the operator for override (PRD 0062).
TOOL_EGRESS_TOKEN_ALLOW = "egress-token-allow"
TOOL_LIST_EGRESS_ROUTES = "list-egress-routes"
TOOLS: tuple[str, ...] = (
TOOL_EGRESS_ALLOW,
TOOL_EGRESS_BLOCK,
TOOL_GITLEAKS_ALLOW,
TOOL_EGRESS_TOKEN_ALLOW,
TOOL_LIST_EGRESS_ROUTES,
)
# The supervise sidecar uses these to query egress's
# introspection endpoint for the `list-egress-routes` MCP
# tool. The hostname + port match egress's docker network
@@ -77,236 +95,77 @@ COMPONENT_FOR_TOOL: dict[str, str] = {
TOOL_EGRESS_BLOCK: "egress",
}
STATUS_APPROVED = "approved"
STATUS_MODIFIED = "modified"
STATUS_REJECTED = "rejected"
STATUSES: tuple[str, ...] = (STATUS_APPROVED, STATUS_MODIFIED, STATUS_REJECTED)
# Operator-initiated audit entries (no tool call). PRD 0014's
# `routes edit <bottle>` verb writes entries with this action.
ACTION_OPERATOR_EDIT = "operator-edit"
QUEUE_DIR_IN_CONTAINER = "/run/supervise/queue"
DB_PATH_IN_CONTAINER = "/run/supervise/bot-bottle.db"
DEFAULT_POLL_INTERVAL_SEC = 0.5
# --- Paths -----------------------------------------------------------------
def bot_bottle_root() -> Path:
return Path.home() / ".bot-bottle"
def queue_dir_for_slug(slug: str) -> Path:
return bot_bottle_root() / "queue" / slug
def audit_dir() -> Path:
return bot_bottle_root() / "audit"
def host_db_path() -> Path:
# Calls bot_bottle_root() through this module's globals so that patches
# on supervise.bot_bottle_root propagate to callers going through
# supervise.host_db_path().
#
# Kept in its own "db" subdirectory (see supervise_types.host_db_path
# for why) — must stay in sync with that copy.
return bot_bottle_root() / "db" / HOST_DB_FILENAME
def audit_log_path(component: str, slug: str) -> Path:
return audit_dir() / f"{component}-{slug}.log"
# --- Dataclasses -----------------------------------------------------------
@dataclass(frozen=True)
class Proposal:
"""One pending tool-call from the agent. The sidecar writes one
of these to the queue dir on a tool call; the operator's TUI
reads them; the sidecar polls for a matching Response."""
id: str
bottle_slug: str
tool: str
proposed_file: str
justification: str
arrival_timestamp: str
current_file_hash: str
@classmethod
def new(
cls,
*,
bottle_slug: str,
tool: str,
proposed_file: str,
justification: str,
current_file_hash: str,
now: datetime | None = None,
) -> "Proposal":
ts = (now or datetime.now(timezone.utc)).isoformat()
return cls(
id=str(uuid.uuid4()),
bottle_slug=bottle_slug,
tool=tool,
proposed_file=proposed_file,
justification=justification,
arrival_timestamp=ts,
current_file_hash=current_file_hash,
)
def to_dict(self) -> dict[str, object]:
return dataclasses.asdict(self)
@classmethod
def from_dict(cls, raw: dict[str, object]) -> "Proposal":
tool = _require_str(raw, "tool")
if tool not in TOOLS:
raise ValueError(f"tool must be one of {TOOLS}; got {tool!r}")
return cls(
id=_require_str(raw, "id"),
bottle_slug=_require_str(raw, "bottle_slug"),
tool=tool,
proposed_file=_require_str(raw, "proposed_file"),
justification=_require_str(raw, "justification"),
arrival_timestamp=_require_str(raw, "arrival_timestamp"),
current_file_hash=_require_str(raw, "current_file_hash"),
)
@dataclass(frozen=True)
class Response:
"""The operator's decision on a proposal. The TUI writes one of
these to the queue dir; the sidecar reads it and returns the
`{status, notes}` pair to the agent's tool call.
`final_file` carries the file content the supervisor will
actually apply: for `approved`, equal to the proposal's
`proposed_file`; for `modified`, the operator's edited version
(the audit diff is current final_file, not current
proposed_file); for `rejected`, None."""
proposal_id: str
status: str
notes: str
final_file: str | None = None
def to_dict(self) -> dict[str, object]:
return dataclasses.asdict(self)
@classmethod
def from_dict(cls, raw: dict[str, object]) -> "Response":
status = _require_str(raw, "status")
if status not in STATUSES:
raise ValueError(
f"response status must be one of {STATUSES}; got {status!r}"
)
final = raw.get("final_file")
if final is not None and not isinstance(final, str):
raise ValueError(
f"final_file must be a string or null; got {type(final).__name__}"
)
return cls(
proposal_id=_require_str(raw, "proposal_id"),
status=status,
notes=_require_str(raw, "notes"),
final_file=final,
)
@dataclass(frozen=True)
class AuditEntry:
"""One row of the per-bottle audit log. JSON-Lines, append-only."""
timestamp: str
bottle_slug: str
component: str
operator_action: str
operator_notes: str
justification: str
diff: str
def to_dict(self) -> dict[str, object]:
return dataclasses.asdict(self)
try:
from .queue_store import QueueStore
from .audit_store import AuditStore
from .store_manager import StoreManager
except ImportError:
# Sidecar bundle: files are flat-copied under /app, not a package.
from queue_store import QueueStore # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
from audit_store import AuditStore # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
from store_manager import StoreManager # type: ignore[import-not-found] # pylint: disable=import-error,no-name-in-module
# --- Queue I/O -------------------------------------------------------------
def _proposal_filename(proposal_id: str) -> str:
return f"{proposal_id}.proposal.json"
def _response_filename(proposal_id: str) -> str:
return f"{proposal_id}.response.json"
def _id_from_proposal_filename(path: Path) -> str | None:
name = path.name
if not name.endswith(".proposal.json"):
return None
return name[: -len(".proposal.json")]
def write_proposal(queue_dir: Path, proposal: Proposal) -> Path:
"""Persist `proposal` as JSON in the queue dir, mode 0o600.
def write_proposal(proposal: Proposal) -> Path:
"""Persist `proposal` in the queue database, mode 0o600.
Directory is created if missing."""
queue_dir.mkdir(parents=True, exist_ok=True)
path = queue_dir / _proposal_filename(proposal.id)
payload = json.dumps(proposal.to_dict(), indent=2) + "\n"
_atomic_write(path, payload, mode=0o600)
return path
return QueueStore(proposal.bottle_slug).write_proposal(proposal)
def read_proposal(queue_dir: Path, proposal_id: str) -> Proposal:
path = queue_dir / _proposal_filename(proposal_id)
with path.open() as f:
raw = json.load(f)
if not isinstance(raw, dict):
raise ValueError(f"{path}: top-level must be an object")
return Proposal.from_dict(raw)
def read_proposal(bottle_slug: str, proposal_id: str) -> Proposal:
return QueueStore(bottle_slug).read_proposal(proposal_id)
def list_pending_proposals(queue_dir: Path) -> list[Proposal]:
"""All proposals in `queue_dir` that do not yet have a matching
response file. Sorted by `arrival_timestamp` so the operator
def list_pending_proposals(bottle_slug: str) -> list[Proposal]:
"""All proposals for `bottle_slug` that do not yet have a matching
response. Sorted by `arrival_timestamp` so the operator
sees the queue FIFO."""
if not queue_dir.is_dir():
return []
out: list[Proposal] = []
for path in sorted(queue_dir.glob("*.proposal.json")):
proposal_id = _id_from_proposal_filename(path)
if proposal_id is None:
continue
if (queue_dir / _response_filename(proposal_id)).exists():
continue
try:
with path.open() as f:
raw = json.load(f)
except (OSError, json.JSONDecodeError):
continue
if not isinstance(raw, dict):
continue
try:
out.append(Proposal.from_dict(raw))
except (KeyError, ValueError):
continue
out.sort(key=lambda p: p.arrival_timestamp)
return out
return QueueStore(bottle_slug).list_pending_proposals()
def write_response(queue_dir: Path, response: Response) -> Path:
queue_dir.mkdir(parents=True, exist_ok=True)
path = queue_dir / _response_filename(response.proposal_id)
payload = json.dumps(response.to_dict(), indent=2) + "\n"
_atomic_write(path, payload, mode=0o600)
return path
def list_all_pending_proposals() -> list[Proposal]:
"""All pending proposals across bottles, sorted FIFO."""
return QueueStore("").list_all_pending_proposals()
def read_response(queue_dir: Path, proposal_id: str) -> Response:
path = queue_dir / _response_filename(proposal_id)
with path.open() as f:
raw = json.load(f)
if not isinstance(raw, dict):
raise ValueError(f"{path}: top-level must be an object")
return Response.from_dict(raw)
def write_response(bottle_slug: str, response: Response) -> Path:
return QueueStore(bottle_slug).write_response(response)
def read_response(bottle_slug: str, proposal_id: str) -> Response:
return QueueStore(bottle_slug).read_response(proposal_id)
def wait_for_response(
queue_dir: Path,
bottle_slug: str,
proposal_id: str,
*,
poll_interval: float = DEFAULT_POLL_INTERVAL_SEC,
@@ -317,90 +176,35 @@ def wait_for_response(
which the wait raises TimeoutError. None waits forever the
natural shape, since the operator's response time is unbounded.
Polls the filesystem so the implementation stays portable and
stdlib-only."""
path = queue_dir / _response_filename(proposal_id)
Polls SQLite so the implementation stays portable and stdlib-only."""
store = QueueStore(bottle_slug)
while True:
if path.exists():
try:
with path.open() as f:
raw = json.load(f)
except (OSError, json.JSONDecodeError):
raw = None
if isinstance(raw, dict):
try:
return Response.from_dict(raw)
except (KeyError, ValueError):
return store.read_response(proposal_id)
except FileNotFoundError:
pass
if deadline is not None and time.monotonic() >= deadline:
raise TimeoutError(f"no response for proposal {proposal_id!r}")
time.sleep(poll_interval)
def archive_proposal(queue_dir: Path, proposal_id: str) -> None:
"""Move both proposal and response files to `<queue_dir>/processed/`.
Idempotent missing files are silently skipped."""
processed = queue_dir / "processed"
processed.mkdir(parents=True, exist_ok=True)
for name in (_proposal_filename(proposal_id), _response_filename(proposal_id)):
src = queue_dir / name
if src.exists():
src.rename(processed / name)
def archive_proposal(bottle_slug: str, proposal_id: str) -> None:
"""Mark both proposal and response rows processed.
Idempotent missing rows are silently skipped."""
QueueStore(bottle_slug).archive_proposal(proposal_id)
# --- Audit log -------------------------------------------------------------
def write_audit_entry(entry: AuditEntry) -> Path:
"""Append `entry` as one JSON-Lines record to the per-bottle
audit log. Acquires an advisory exclusive lock so concurrent
writers don't interleave bytes."""
path = audit_log_path(entry.component, entry.bottle_slug)
path.parent.mkdir(parents=True, exist_ok=True)
line = json.dumps(entry.to_dict(), sort_keys=False) + "\n"
fd = os.open(path, os.O_WRONLY | os.O_APPEND | os.O_CREAT, 0o600)
try:
_try_flock(fd)
try:
os.write(fd, line.encode("utf-8"))
finally:
_try_funlock(fd)
finally:
os.close(fd)
return path
"""Append `entry` to the host supervise audit table."""
return AuditStore().write_audit_entry(entry)
def read_audit_entries(component: str, slug: str) -> list[AuditEntry]:
"""Load all audit entries for the given component+slug. Empty
list if the log doesn't exist."""
path = audit_log_path(component, slug)
if not path.is_file():
return []
out: list[AuditEntry] = []
with path.open() as f:
for raw_line in f:
raw_line = raw_line.strip()
if not raw_line:
continue
try:
raw = json.loads(raw_line)
except json.JSONDecodeError:
continue
if not isinstance(raw, dict):
continue
try:
out.append(AuditEntry(
timestamp=_require_str(raw, "timestamp"),
bottle_slug=_require_str(raw, "bottle_slug"),
component=_require_str(raw, "component"),
operator_action=_require_str(raw, "operator_action"),
operator_notes=_require_str(raw, "operator_notes"),
justification=_require_str(raw, "justification"),
diff=_require_str(raw, "diff"),
))
except ValueError:
continue
return out
"""Load all audit entries for the given component+slug."""
return AuditStore().read_audit_entries(component, slug)
# --- Diff rendering --------------------------------------------------------
@@ -433,89 +237,47 @@ def sha256_hex(content: str) -> str:
class SupervisePlan:
"""Output of Supervise.prepare; consumed by .start.
`queue_dir` is the host directory bind-mounted into the sidecar
at /run/supervise/queue. `internal_network` is empty at prepare
time; the backend's launch step fills it via dataclasses.replace
before calling .start."""
`db_path` is the host database bind-mounted into the sidecar at
/run/supervise/bot-bottle.db. `internal_network` is empty at
prepare time; the backend's launch step fills it via
dataclasses.replace before calling .start."""
slug: str
queue_dir: Path
db_path: Path
internal_network: str = ""
class Supervise(ABC):
"""Per-bottle supervise sidecar. Encapsulates the host-side
prepare (queue dir staging); the sidecar's start/stop lifecycle
is backend-specific."""
"""Per-bottle supervise sidecar. Encapsulates host-side database
staging; the sidecar's start/stop lifecycle is backend-specific."""
def prepare(
self,
slug: str,
stage_dir: Path,
) -> SupervisePlan:
"""Stage the per-bottle queue dir on the host. Returns the
plan; `internal_network` must be set by the launch step before
.start runs."""
"""Stage the host database. Returns the plan; `internal_network`
must be set by the launch step before .start runs."""
del stage_dir
queue_dir = queue_dir_for_slug(slug)
queue_dir.mkdir(parents=True, exist_ok=True)
mgr = StoreManager.instance()
mgr.migrate()
return SupervisePlan(
slug=slug,
queue_dir=queue_dir,
db_path=mgr.db_path,
)
# --- Helpers ---------------------------------------------------------------
def _require_str(raw: dict[str, object], key: str) -> str:
value = raw.get(key)
if not isinstance(value, str):
raise ValueError(f"missing or non-string field {key!r}")
return value
def _atomic_write(path: Path, content: str, *, mode: int) -> None:
"""Atomic: write to a sibling tmp file, fsync, rename."""
tmp = path.with_suffix(path.suffix + ".tmp")
fd = os.open(tmp, os.O_WRONLY | os.O_CREAT | os.O_TRUNC, mode)
try:
os.write(fd, content.encode("utf-8"))
os.fsync(fd)
finally:
os.close(fd)
os.replace(tmp, path)
try:
import fcntl as _fcntl
def _try_flock(fd: int) -> None: # type: ignore[reportRedeclaration]
try:
_fcntl.flock(fd, _fcntl.LOCK_EX)
except OSError:
pass
def _try_funlock(fd: int) -> None: # type: ignore[reportRedeclaration]
try:
_fcntl.flock(fd, _fcntl.LOCK_UN)
except OSError:
pass
except ImportError: # pragma: no cover — Windows path
def _try_flock(fd: int) -> None: # noqa: F841 — Windows fallback
return None
def _try_funlock(fd: int) -> None: # noqa: F841 — Windows fallback
return None
__all__ = [
"ACTION_OPERATOR_EDIT",
"AuditEntry",
"AuditStore",
"COMPONENT_FOR_TOOL",
"DEFAULT_POLL_INTERVAL_SEC",
"DB_PATH_IN_CONTAINER",
"Proposal",
"QUEUE_DIR_IN_CONTAINER",
"QueueStore",
"Response",
"StoreManager",
"STATUSES",
"STATUS_APPROVED",
"STATUS_MODIFIED",
@@ -536,8 +298,9 @@ __all__ = [
"audit_dir",
"audit_log_path",
"bot_bottle_root",
"host_db_path",
"list_pending_proposals",
"queue_dir_for_slug",
"list_all_pending_proposals",
"read_audit_entries",
"read_proposal",
"read_response",
+9 -17
View File
@@ -7,14 +7,13 @@ config changes when stuck. The tools are `egress-allow`,
Each queued tool call:
1. Validates the proposed file syntactically.
2. Writes a Proposal to /run/supervise/queue/ (bind-mounted from
the host's ~/.bot-bottle/queue/<slug>/).
3. Blocks polling for a matching Response file.
2. Writes a Proposal to the host SQLite database.
3. Blocks polling for a matching Response row.
4. Returns the operator's `{status, notes}` to the agent.
The bottle slug arrives via SUPERVISE_BOTTLE_SLUG env (stamped at
container creation by the backend's start step). The queue dir comes
from SUPERVISE_QUEUE_DIR (default `/run/supervise/queue`).
container creation by the backend's start step). SUPERVISE_DB_PATH
points at the bind-mounted host database.
Speaks MCP over HTTP+JSON-RPC. Methods handled:
@@ -42,7 +41,6 @@ import typing
import urllib.error
import urllib.request
from dataclasses import dataclass
from pathlib import Path
try:
# Same-directory imports inside the bundle container; these files are
@@ -277,7 +275,6 @@ def validate_proposed_file(tool: str, content: str) -> None:
@dataclass(frozen=True)
class ServerConfig:
bottle_slug: str
queue_dir: Path
response_timeout_seconds: float = DEFAULT_RESPONSE_TIMEOUT_SECONDS
@@ -376,7 +373,7 @@ def handle_tools_call(
current_file_hash=_sv.sha256_hex(proposed_file),
)
try:
_sv.write_proposal(config.queue_dir, proposal)
_sv.write_proposal(proposal)
except OSError as e:
raise _RpcInternalError(f"failed to write proposal to queue: {e}") from e
sys.stderr.write(
@@ -387,7 +384,7 @@ def handle_tools_call(
deadline = time.monotonic() + config.response_timeout_seconds
try:
response = _sv.wait_for_response(
config.queue_dir,
config.bottle_slug,
proposal.id,
poll_interval=MIN_RESPONSE_POLL_INTERVAL_SECONDS,
deadline=deadline,
@@ -399,7 +396,7 @@ def handle_tools_call(
"isError": False,
}
try:
_sv.archive_proposal(config.queue_dir, proposal.id)
_sv.archive_proposal(config.bottle_slug, proposal.id)
except OSError as e:
raise _RpcInternalError(f"failed to archive proposal: {e}") from e
@@ -539,7 +536,7 @@ class MCPHandler(http.server.BaseHTTPRequestHandler):
class MCPServer(socketserver.ThreadingMixIn, http.server.HTTPServer):
allow_reuse_address = True
daemon_threads = True
config: ServerConfig = ServerConfig(bottle_slug="", queue_dir=Path())
config: ServerConfig = ServerConfig(bottle_slug="")
# --- Entry point -----------------------------------------------------------
@@ -548,21 +545,18 @@ class MCPServer(socketserver.ThreadingMixIn, http.server.HTTPServer):
def serve(
*,
bottle_slug: str,
queue_dir: Path,
port: int = _sv.SUPERVISE_PORT,
bind: str = "0.0.0.0",
response_timeout_seconds: float = DEFAULT_RESPONSE_TIMEOUT_SECONDS,
) -> typing.NoReturn:
queue_dir.mkdir(parents=True, exist_ok=True)
server = MCPServer((bind, port), MCPHandler)
server.config = ServerConfig(
bottle_slug=bottle_slug,
queue_dir=queue_dir,
response_timeout_seconds=response_timeout_seconds,
)
sys.stderr.write(
f"supervise listening on {bind}:{port}; "
f"slug={bottle_slug!r}; queue={queue_dir}; "
f"slug={bottle_slug!r}; "
f"tools: {', '.join(t['name'] for t in TOOL_DEFINITIONS)}\n" # type: ignore[arg-type]
)
sys.stderr.flush()
@@ -581,7 +575,6 @@ def main(argv: list[str]) -> int:
if not bottle_slug:
sys.stderr.write("supervise: SUPERVISE_BOTTLE_SLUG env is unset\n")
return 2
queue_dir = Path(os.environ.get("SUPERVISE_QUEUE_DIR", _sv.QUEUE_DIR_IN_CONTAINER))
port = int(os.environ.get("SUPERVISE_PORT", str(_sv.SUPERVISE_PORT)))
bind = os.environ.get("SUPERVISE_BIND", "0.0.0.0")
try:
@@ -591,7 +584,6 @@ def main(argv: list[str]) -> int:
return 2
serve(
bottle_slug=bottle_slug,
queue_dir=queue_dir,
port=port,
bind=bind,
response_timeout_seconds=response_timeout_seconds,
+189
View File
@@ -0,0 +1,189 @@
"""Shared types and path helpers for supervise stores (PRD 0013).
Extracted from supervise.py so queue_store and audit_store can import
Proposal, Response, AuditEntry, and host_db_path without creating a
circular import (supervise imports from queue_store/audit_store and
vice-versa).
Patching bot_bottle_root on this module propagates through host_db_path
because host_db_path looks up bot_bottle_root via sys.modules[__name__]
at call time rather than capturing it at import time.
"""
from __future__ import annotations
import dataclasses
import sys
import uuid
from dataclasses import dataclass
from datetime import datetime, timezone
from pathlib import Path
HOST_DB_FILENAME = "bot-bottle.db"
TOOL_EGRESS_BLOCK = "egress-block"
TOOL_EGRESS_ALLOW = "egress-allow"
TOOL_GITLEAKS_ALLOW = "gitleaks-allow"
TOOL_EGRESS_TOKEN_ALLOW = "egress-token-allow"
TOOL_LIST_EGRESS_ROUTES = "list-egress-routes"
TOOLS: tuple[str, ...] = (
TOOL_EGRESS_ALLOW,
TOOL_EGRESS_BLOCK,
TOOL_GITLEAKS_ALLOW,
TOOL_EGRESS_TOKEN_ALLOW,
TOOL_LIST_EGRESS_ROUTES,
)
STATUS_APPROVED = "approved"
STATUS_MODIFIED = "modified"
STATUS_REJECTED = "rejected"
STATUSES: tuple[str, ...] = (STATUS_APPROVED, STATUS_MODIFIED, STATUS_REJECTED)
ACTION_OPERATOR_EDIT = "operator-edit"
def bot_bottle_root() -> Path:
return Path.home() / ".bot-bottle"
def host_db_path() -> Path:
# Look up bot_bottle_root through this module's live namespace so that
# monkey-patches on supervise_types.bot_bottle_root take effect.
#
# Lives in its own "db" subdirectory, not directly under
# bot_bottle_root(), so backends that can only bind-mount
# directories (macos_container's `container run --mount`, which
# rejects file sources with "is not a directory") can share this
# one file with a sidecar without also exposing bot_bottle_root()'s
# other contents (git-gate keys, per-bottle state, etc.).
return sys.modules[__name__].bot_bottle_root() / "db" / HOST_DB_FILENAME
def _require_str(raw: dict[str, object], key: str) -> str:
value = raw.get(key)
if not isinstance(value, str):
raise ValueError(f"missing or non-string field {key!r}")
return value
@dataclass(frozen=True)
class Proposal:
"""One pending tool-call from the agent."""
id: str
bottle_slug: str
tool: str
proposed_file: str
justification: str
arrival_timestamp: str
current_file_hash: str
@classmethod
def new(
cls,
*,
bottle_slug: str,
tool: str,
proposed_file: str,
justification: str,
current_file_hash: str,
now: datetime | None = None,
) -> "Proposal":
ts = (now or datetime.now(timezone.utc)).isoformat()
return cls(
id=str(uuid.uuid4()),
bottle_slug=bottle_slug,
tool=tool,
proposed_file=proposed_file,
justification=justification,
arrival_timestamp=ts,
current_file_hash=current_file_hash,
)
def to_dict(self) -> dict[str, object]:
return dataclasses.asdict(self)
@classmethod
def from_dict(cls, raw: dict[str, object]) -> "Proposal":
tool = _require_str(raw, "tool")
if tool not in TOOLS:
raise ValueError(f"tool must be one of {TOOLS}; got {tool!r}")
return cls(
id=_require_str(raw, "id"),
bottle_slug=_require_str(raw, "bottle_slug"),
tool=tool,
proposed_file=_require_str(raw, "proposed_file"),
justification=_require_str(raw, "justification"),
arrival_timestamp=_require_str(raw, "arrival_timestamp"),
current_file_hash=_require_str(raw, "current_file_hash"),
)
@dataclass(frozen=True)
class Response:
"""The operator's decision on a proposal."""
proposal_id: str
status: str
notes: str
final_file: str | None = None
def to_dict(self) -> dict[str, object]:
return dataclasses.asdict(self)
@classmethod
def from_dict(cls, raw: dict[str, object]) -> "Response":
status = _require_str(raw, "status")
if status not in STATUSES:
raise ValueError(
f"response status must be one of {STATUSES}; got {status!r}"
)
final = raw.get("final_file")
if final is not None and not isinstance(final, str):
raise ValueError(
f"final_file must be a string or null; got {type(final).__name__}"
)
return cls(
proposal_id=_require_str(raw, "proposal_id"),
status=status,
notes=_require_str(raw, "notes"),
final_file=final,
)
@dataclass(frozen=True)
class AuditEntry:
"""One row of the per-bottle audit log."""
timestamp: str
bottle_slug: str
component: str
operator_action: str
operator_notes: str
justification: str
diff: str
def to_dict(self) -> dict[str, object]:
return dataclasses.asdict(self)
__all__ = [
"ACTION_OPERATOR_EDIT",
"AuditEntry",
"HOST_DB_FILENAME",
"Proposal",
"Response",
"STATUSES",
"STATUS_APPROVED",
"STATUS_MODIFIED",
"STATUS_REJECTED",
"TOOLS",
"TOOL_EGRESS_ALLOW",
"TOOL_EGRESS_BLOCK",
"TOOL_EGRESS_TOKEN_ALLOW",
"TOOL_GITLEAKS_ALLOW",
"TOOL_LIST_EGRESS_ROUTES",
"bot_bottle_root",
"host_db_path",
]
+135
View File
@@ -0,0 +1,135 @@
# PRD 0067: SQLite local storage
- **Status:** Active
- **Author:** codex
- **Created:** 2026-07-01
- **Issue:** #319
## Summary
Add a small stdlib SQLite storage layer for bot-bottle host runtime state,
starting with the supervise queue and audit log. This replaces scattered JSON
queue files and JSONL audit logs with structured tables while preserving the
existing public supervise helper functions and sidecar queue mount contract.
## Problem
Bot-bottle currently stores supervise proposals and responses as individual JSON
files under `~/.bot-bottle/queue/<slug>/`, and audit entries as JSONL files
under `~/.bot-bottle/audit/`. That worked for the original interactive TUI, but
new forge-native orchestration needs durable, queryable local state for queues,
audit trails, watchdogs, and lifecycle records. PR #318 started introducing
SQLite-shaped boilerplate for forge state; the storage foundation should live in
its own PR so forge work can build on the shared runtime store instead of adding
one-off persistence.
## Goals / Success Criteria
1. Supervise proposals and responses are persisted through SQLite.
2. Audit entries are persisted through SQLite.
3. Supervise queue helpers use the bottle slug / queue key instead of a queue
directory path.
4. The sidecar receives the host database mount across docker, smolmachines,
and macOS-container backends.
5. The implementation stays stdlib-only.
6. Schema migrations use a `PRAGMA user_version` runner — no third-party deps.
7. Unit tests cover queue round-trips, pending discovery, response waits,
archive semantics, audit round-trips, and path creation.
## Non-goals
- Migrating old JSON queue files or JSONL audit logs.
- Adding forge orchestration state tables.
- Adding egress metering or budget tables.
- Changing the supervise TUI workflow or remediation behavior.
- Introducing a third-party ORM or migration library.
## Design
### Database locations
Queue and audit state use the host-level local database:
```text
~/.bot-bottle/bot-bottle.db
```
The supervise sidecar receives that database as a writable bind mount at
`/run/supervise/bot-bottle.db` and gets the path through `SUPERVISE_DB_PATH`.
No per-slug queue directory is mounted into the sidecar. This creates the shared
host database that later forge/native lifecycle work can extend in separate
PRDs.
### Tables
`supervise_proposals` lives in the host database:
```sql
CREATE TABLE supervise_proposals (
queue_key TEXT NOT NULL,
id TEXT NOT NULL,
bottle_slug TEXT NOT NULL,
tool TEXT NOT NULL,
proposed_file TEXT NOT NULL,
justification TEXT NOT NULL,
arrival_timestamp TEXT NOT NULL,
current_file_hash TEXT NOT NULL,
archived INTEGER NOT NULL DEFAULT 0,
PRIMARY KEY (queue_key, id)
);
```
`supervise_responses` lives in the host database:
```sql
CREATE TABLE supervise_responses (
queue_key TEXT NOT NULL,
proposal_id TEXT NOT NULL,
status TEXT NOT NULL,
notes TEXT NOT NULL,
final_file TEXT,
archived INTEGER NOT NULL DEFAULT 0,
PRIMARY KEY (queue_key, proposal_id)
);
```
`supervise_audit_entries` lives in the host database:
```sql
CREATE TABLE supervise_audit_entries (
id INTEGER PRIMARY KEY AUTOINCREMENT,
timestamp TEXT NOT NULL,
bottle_slug TEXT NOT NULL,
component TEXT NOT NULL,
operator_action TEXT NOT NULL,
operator_notes TEXT NOT NULL,
justification TEXT NOT NULL,
diff TEXT NOT NULL
);
```
### Compatibility
The queue helpers take a bottle slug / queue key and perform equivalent
operations against `~/.bot-bottle/bot-bottle.db`:
- `list_pending_proposals` returns non-archived proposals without a non-archived
response, sorted by arrival time.
- `archive_proposal` marks matching proposal/response rows archived instead of
moving files into `processed/`.
- `wait_for_response` keeps the current polling behavior but polls SQLite.
The old audit path helpers (`audit_dir`, `audit_log_path`) stay available for
compatibility. `audit_log_path` no longer describes the active storage location;
callers should use `read_audit_entries`.
## Implementation chunks
1. Add SQLite store helpers for supervise queue and audit state.
2. Rewire `bot_bottle.supervise` queue/audit functions to the store.
3. Update supervise CLI discovery tests and queue/audit unit tests.
4. Run unit tests, pyright, and pylint for touched modules.
## Open questions
None.
+227
View File
@@ -0,0 +1,227 @@
# PRD prd-new: smolmachines backend on Linux
- **Status:** Draft
- **Author:** Claude
- **Created:** 2026-06-25
- **Issue:** #283
## Summary
Make the `smolmachines` backend (PRD 0023) runnable on Linux, not
just macOS. `smolvm` already supports Linux via KVM (`/dev/kvm`);
the gap is entirely in bot-bottle's host-side glue, which hard-codes
macOS assumptions in three places:
1. **Preflight** only checks that `smolvm` is on `PATH` — it never
checks the Linux KVM prerequisite, so a misconfigured host fails
deep in the launch flow with an opaque `smolvm` error.
2. **The TSI allowlist enforcement** (`force_allowlist`) — the
security property that confines the agent VM to its sidecar
bundle's `/32`**no-ops on Linux today, failing _open_**. The
smolvm state-DB path it patches is hard-coded to macOS's
`~/Library/Application Support/...`.
3. **Per-bottle loopback scoping** (`allocate`) returns the shared
`127.0.0.1` on Linux, which would let the agent VM reach every
service on host loopback — a downgrade from the per-bottle alias
isolation macOS gets.
This PRD closes all three so a bottle launched with
`BOT_BOTTLE_BACKEND=smolmachines` on Linux gets the same isolation
guarantee it gets on macOS, and documents the Linux/NixOS host
setup. The primary validation target is NixOS, but the changes are
distro-agnostic.
## Problem
The smolmachines backend runs each bottle's agent inside a libkrun
microVM via `smolvm`, with egress confined by TSI's `--allow-cidr`
allowlist set to a single `/32` — the sidecar bundle's loopback
address. Everything else (host loopback, LAN, internet) is denied at
the VMM layer. That security property is the entire reason the
backend exists.
libkrun runs on Hypervisor.framework (macOS) **and** KVM (Linux), and
`smolvm` ships Linux x86_64 / aarch64 builds that require `/dev/kvm`.
So the microVM layer already works on Linux. What does not work is
bot-bottle's host integration, which PRD 0023 explicitly scoped to
macOS-only for v1. Three concrete blockers:
- **No KVM preflight.** On a Linux host without `/dev/kvm` (kernel
module not loaded) or without access to it (user not in the `kvm`
group), the failure surfaces as a cryptic `smolvm` non-zero exit
mid-launch instead of an actionable message.
- **TSI enforcement fails open on Linux.** `force_allowlist`
early-returns on non-macOS. It exists because `smolvm` 0.8.0
silently drops `--allow-cidr` when combined with `--from`, so the
allowlist has to be patched into smolvm's persisted state DB before
`machine start`. On Linux that patch never runs **and** the DB path
is the macOS path, so the booted VM's TSI allowlist is whatever
smolvm defaulted to — potentially all of `127.0.0.0/8`. That is the
exact sandbox-escape the backend is supposed to prevent.
- **No per-bottle loopback isolation on Linux.** `allocate` returns
`127.0.0.1` on Linux. Even with a correct allowlist, `127.0.0.1/32`
is shared by every service on host loopback, so the agent could
reach other bottles' published ports and host services. On macOS
this is solved with per-bottle `127.0.0.16..31` aliases added via
`sudo ifconfig lo0 alias`. On Linux the whole `127.0.0.0/8` is
already routed to `lo`, so docker can publish to `127.0.0.<N>`
with **no `ifconfig`/sudo step at all** — the isolation is actually
cheaper to achieve than on macOS.
## Goals / Success Criteria
- `BOT_BOTTLE_BACKEND=smolmachines ./cli.py start <agent>` launches,
runs, and tears down a bottle on a Linux host with `/dev/kvm`.
- The TSI allowlist is enforced on Linux: PRD 0022's
`tests/integration/test_sandbox_escape.py` passes against
`BOT_BOTTLE_BACKEND=smolmachines` on Linux (the acceptance gate).
- Each Linux bottle is scoped to its own `127.0.0.<N>/32`, matching
the macOS per-bottle isolation property.
- A clear, actionable preflight error when `/dev/kvm` is missing or
inaccessible, with remediation (load `kvm-intel`/`kvm-amd`, join the
`kvm` group).
- **Fail-closed:** if bot-bottle cannot positively confirm the TSI
allowlist was persisted for a machine (DB missing, row missing,
patch didn't take), it `die()`s before `machine start` rather than
booting a VM with an unverified allowlist.
- macOS behavior is unchanged.
- README documents Linux + NixOS host setup.
## Non-goals
- Rootless / non-KVM fallbacks (e.g. software emulation). Linux
smolmachines requires `/dev/kvm`, full stop.
- Removing Docker as a host dependency — the sidecar bundle and
image-build pipeline still use Docker on Linux, same as macOS.
- Auto-installing `smolvm` or configuring KVM on the operator's
behalf. Preflight reports; the operator remediates.
- Nested-virtualization tuning for running the runner itself inside a
VM (documented as a caveat, not solved here).
## Design
### Platform detection
Reuse the existing `platform.system()` check already in
`loopback_alias.py` (`_is_macos()`). "Linux" is "not macOS" for every
branch below; no new third-platform path.
### Preflight: KVM gate (`util.smolmachines_preflight`)
After the existing `smolvm`-on-`PATH` check, add a Linux-only gate:
- `/dev/kvm` must exist → else `die()` with "enable KVM
(`kvm-intel`/`kvm-amd` kernel module)".
- `/dev/kvm` must be readable + writable by the current user
(`os.access(..., R_OK | W_OK)`) → else `die()` with "add your user
to the `kvm` group (and re-login)".
macOS is unaffected (Hypervisor.framework needs no device node).
### smolvm state-DB path (platform-aware)
`loopback_alias._SMOLVM_DB_PATH` becomes platform-derived:
- macOS: `~/Library/Application Support/smolvm/server/smolvm.db`
(unchanged).
- Linux: `$XDG_DATA_HOME/smolvm/server/smolvm.db`, defaulting to
`~/.local/share/smolvm/server/smolvm.db`.
> **Verification note:** the Linux DB location is inferred from
> smolvm's documented `~/.local/share` install layout and the XDG
> base-dir spec. It must be confirmed on a real Linux smolvm install;
> if smolvm uses a different path or schema, the fail-closed check
> below turns that into a clear `die()` at launch rather than a silent
> escape.
### TSI enforcement: cross-platform + fail-closed (`force_allowlist`)
Rework `force_allowlist(machine_name, allowed_cidrs)` to run on
**both** platforms and to fail closed:
1. Resolve the state DB; if the file is missing, `die()` (cannot
confirm enforcement → refuse to launch).
2. Read the machine's persisted row; if the row is missing, `die()`.
3. If the row's `allowed_cidrs` already equals the requested list
(e.g. a newer `smolvm` that honors `--allow-cidr` at create), do
nothing — no write.
4. Otherwise patch `allowed_cidrs` (the existing BLOB-encoded write)
and re-read.
5. If, after the patch, `allowed_cidrs` still does not equal the
requested list, `die()`.
This is robust across smolvm versions: it works whether `--allow-cidr`
is silently dropped (0.8.0) or honored (newer), and it never boots a
VM whose persisted allowlist it could not confirm. It is a strict
improvement on macOS too (today's code writes unconditionally and
never verifies).
> The persisted-row check confirms our write took, not that smolvm's
> runtime TSI enforces it. The runtime guarantee is covered by the
> sandbox-escape acceptance test; the persisted check is the cheap
> fail-closed guard at launch.
### Per-bottle loopback scoping on Linux (`allocate`)
`allocate` runs the same docker-state-driven allocation on Linux as on
macOS (`_allocate_locked`, the file lock, and `_aliases_in_use` via
`docker inspect` are all already cross-platform). The only macOS-only
step, `ensure_pool` (the `sudo ifconfig lo0 alias` dance), stays
macOS-only: on Linux `127.0.0.0/8` is already loopback, so docker can
publish bundle ports directly on `127.0.0.<N>` with no setup.
Net effect: Linux bottles get per-bottle `127.0.0.16..31/32` scoping
identical to macOS, without sudo.
### Launch flow
`launch.py` needs no structural change — `_allocate_resources` already
calls `ensure_pool()` (now a Linux no-op) then `allocate()` (now
per-bottle on Linux), and `_launch_vm` already calls
`force_allowlist()` (now active on Linux). Only the macOS-specific
docstrings are updated to describe the cross-platform behavior.
## Implementation chunks
1. **Preflight KVM gate**`util.smolmachines_preflight` +
unit tests for the missing-device and no-access branches.
2. **Platform-aware DB path + fail-closed `force_allowlist`**
`loopback_alias.py`; update/extend `TestForceAllowlist`.
3. **Cross-platform `allocate`** — drop the Linux early-return; update
`TestAllocate` / `TestAllocateLock` for the new Linux behavior.
4. **Docstring + comment cleanup** in `launch.py` and module headers.
5. **Docs** — README requirements + a Linux/NixOS host-setup section.
## Testing Strategy
- **Unit (CI, any OS):** the suite mocks `platform.system()` /
`subprocess` and patches `_SMOLVM_DB_PATH`, so the new Linux
branches are testable on the macOS/Linux CI runner without `smolvm`
or KVM. Covers: KVM preflight branches, fail-closed `force_allowlist`
(DB missing, row missing, patch-doesn't-take), per-bottle Linux
allocation + locking, platform-derived DB path.
- **Integration (Linux host with KVM — the acceptance gate):**
`tests/integration/test_sandbox_escape.py` against
`BOT_BOTTLE_BACKEND=smolmachines`. This cannot run on the macOS dev
box and must be executed on NixOS before merge.
## Open questions / verification pending
- **Confirm the Linux smolvm state-DB path and schema** on a real
install (the `~/.local/share/...` inference above).
- **Confirm whether the current smolvm Linux build still drops
`--allow-cidr` with `--from`** (the 0.8.0 bug). The fail-closed
design handles either answer, but knowing lets us drop the DB patch
if upstream fixed it.
- **Confirm docker publishing to `127.0.0.<N>` on Linux** behaves as
expected end-to-end with TSI (high confidence; standard loopback
behavior, but unverified on the target host).
## References
- PRD 0023 — smolmachines bottle backend (macOS v1).
- PRD 0022 — `test_sandbox_escape.py` acceptance gate.
- PRD 0024 — sidecar bundle image.
- smolvm: https://github.com/smol-machines/smolvm
+3 -3
View File
@@ -81,10 +81,10 @@ class TestSandboxEscape(unittest.TestCase):
# those are missing rather than die-ing inside backend.prepare.
backend_name = os.environ.get("BOT_BOTTLE_BACKEND", "docker")
if backend_name == "smolmachines":
if sys.platform != "darwin":
if sys.platform not in ("darwin", "linux"):
raise unittest.SkipTest(
"BOT_BOTTLE_BACKEND=smolmachines is macOS-only in "
"v1 (libkrun TSI)"
f"BOT_BOTTLE_BACKEND=smolmachines is not supported "
f"on {sys.platform} (macOS and Linux only)"
)
if shutil.which("smolvm") is None:
raise unittest.SkipTest(
+12 -4
View File
@@ -21,7 +21,7 @@ security properties the design pivot was about:
bind-address mitigation is what closes TSI's port-granularity
gap.
Gated on macOS + smolvm + docker + not GITEA_ACTIONS the
Gated on macOS/Linux + smolvm + docker + not GITEA_ACTIONS the
runner can't host libkrun-backed VMs."""
from __future__ import annotations
@@ -65,8 +65,8 @@ def _minimal_manifest() -> ManifestIndex:
@skip_unless_docker()
@unittest.skipUnless(
platform.system() == "Darwin",
"smolvm is macOS-only for v1; Linux+KVM path is a future PRD",
platform.system() in ("Darwin", "Linux"),
"smolvm requires macOS or Linux",
)
@unittest.skipUnless(
_smolvm_available(),
@@ -141,7 +141,15 @@ class TestSmolmachinesLaunch(unittest.TestCase):
proxies = [line.strip() for line in r.stdout.splitlines()]
self.assertEqual(2, len(proxies), proxies)
self.assertEqual(proxies[0], proxies[1], proxies)
self.assertTrue(proxies[0].startswith("http://127."), proxies[0])
# macOS: proxy binds to the per-bottle loopback alias (127.x.x.x) so
# TSI can intercept guest connections to it. Linux: the guest kernel
# routes 127.0.0.0/8 to its own loopback (TSI never sees those), so
# the proxy instead binds to the per-bottle bridge gateway (192.168.x.1)
# which routes via eth0 and is intercepted by TSI normally.
self.assertRegex(
proxies[0], r"^http://\d+\.\d+\.\d+\.\d+:\d+$",
"expected proxy URL to be an http://IP:port address",
)
r = self.bottle.exec(
"curl -fsS --max-time 20 https://example.com >/dev/null && echo OK"
@@ -2,7 +2,7 @@
exercised against the real binary.
The full machine-lifecycle round trip (create start exec
delete) is gated behind macOS + Darwin platform check and lives
delete) is gated behind macOS/Linux platform check and lives
in chunk 2d's smoke. This file just verifies `is_available()`
correctly reports presence and `_smolvm()` can run a no-op
subcommand without errors enough to flag wrapper drift if
@@ -23,8 +23,8 @@ from bot_bottle.backend.smolmachines.smolvm import is_available
"skipped under act_runner: smolvm not installed on the runner",
)
@unittest.skipUnless(
platform.system() == "Darwin",
"smolvm is macOS-only for v1; Linux+KVM path is a future PRD",
platform.system() in ("Darwin", "Linux"),
"smolvm requires macOS or Linux",
)
@unittest.skipUnless(
is_available(),
+4 -1
View File
@@ -35,7 +35,10 @@ class TestAgentProviderRuntime(unittest.TestCase):
)
config = Path(tmp, "codex-config.toml").read_text()
self.assertEqual("codex", plan.template)
self.assertEqual("codex", plan.command)
self.assertEqual(
"/home/node/.codex/packages/standalone/current/bin/codex",
plan.command,
)
self.assertEqual("read_prompt_file", plan.prompt_mode)
self.assertEqual("/tmp/Dockerfile.codex", plan.dockerfile)
self.assertEqual(
+17
View File
@@ -12,11 +12,19 @@ from unittest.mock import patch
import bot_bottle.cli as climod
from bot_bottle.cli import main
from bot_bottle.db_store import DbStore
from bot_bottle.log import Die
from bot_bottle.manifest import ManifestError
from bot_bottle.store_manager import StoreManager
class TestMainDispatch(unittest.TestCase):
def setUp(self) -> None:
patcher = patch.object(DbStore, "is_migrated", return_value=True)
self._mock_check = patcher.start()
self.addCleanup(patcher.stop)
self.addCleanup(StoreManager.reset)
def test_no_args_prints_usage_returns_2(self) -> None:
with patch("sys.stderr", io.StringIO()):
self.assertEqual(2, main([]))
@@ -56,6 +64,15 @@ class TestMainDispatch(unittest.TestCase):
main(["x", "a", "b"])
self.assertEqual([["a", "b"]], seen)
def test_missing_env_var_error_maps_to_1(self) -> None:
from bot_bottle.errors import MissingEnvVarError
def boom(_rest: list[str]) -> int:
raise MissingEnvVarError("MY_VAR", "MY_VAR is not set")
with patch.dict(climod.COMMANDS, {"x": boom}), patch("sys.stderr", io.StringIO()):
self.assertEqual(1, main(["x"]))
def test_manifest_error_maps_to_1(self) -> None:
def boom(_rest: list[str]) -> int:
raise ManifestError("bad manifest")
+239
View File
@@ -0,0 +1,239 @@
"""Unit: `cli.py start --headless` non-interactive launch path.
Headless is the keystone for orchestrators, CI, and webhook
dispatch: agent/bottles/label come from flags + manifest defaults, no
TUI selectors fire, and the preflight y/N is auto-confirmed
(`assume_yes=True`). All actual launch work is stubbed so no container
is created.
"""
from __future__ import annotations
import io
import os
import unittest
from unittest.mock import MagicMock, patch
import bot_bottle.cli.start as start_mod
import bot_bottle.cli.tui as tui_mod
from bot_bottle.backend import ActiveAgent
from bot_bottle.log import Die
from bot_bottle.manifest import ManifestError
def _make_manifest(
agent_names: list[str],
bottle_names: list[str] | None = None,
agent_bottle: str = "",
):
manifest = MagicMock()
manifest.agents = {name: MagicMock(bottle=agent_bottle) for name in agent_names}
manifest.all_agent_names = sorted(agent_names)
manifest.all_bottle_names = sorted(bottle_names or [])
manifest.home_md = None # eager mode so _peek_agent_bottle uses agents dict
manifest.require_agent = MagicMock(return_value=None)
return manifest
def _active_agent(slug: str) -> ActiveAgent:
return ActiveAgent(
backend_name="docker",
slug=slug,
agent_name="demo",
started_at="2026-01-01T00:00:00+00:00",
services=(),
)
class TestCmdStartHeadless(unittest.TestCase):
"""Drive `cmd_start --headless` with launch + TUI stubbed out."""
def setUp(self):
self._manifest = _make_manifest(
["researcher", "implementer"], ["claude", "dev"], agent_bottle="claude"
)
patch(
"bot_bottle.cli.start.ManifestIndex.resolve",
return_value=self._manifest,
).start()
self._launch_mock = patch(
"bot_bottle.cli.start._launch_bottle", return_value=0
).start()
# No bottles running by default → no label collision.
patch(
"bot_bottle.cli.start.enumerate_active_agents", return_value=[]
).start()
# If any TUI picker fires in headless mode, that's a bug.
self._agent_picker = patch.object(tui_mod, "filter_select").start()
self._bottle_picker = patch.object(tui_mod, "filter_multiselect").start()
self._modal = patch.object(tui_mod, "name_color_modal").start()
patch.dict(os.environ, {}, clear=False).start()
os.environ.pop("BOT_BOTTLE_BACKEND", None)
self.addCleanup(patch.stopall)
def _spec(self):
self._launch_mock.assert_called_once()
return self._launch_mock.call_args[0][0]
# -- no TUI in headless --------------------------------------------
def test_headless_fires_no_pickers(self):
rc = start_mod.cmd_start(
["--headless", "researcher", "--bottle", "claude", "--prompt", "Do it"]
)
self.assertEqual(0, rc)
self._agent_picker.assert_not_called()
self._bottle_picker.assert_not_called()
self._modal.assert_not_called()
def test_headless_assume_yes_forwarded(self):
start_mod.cmd_start(
["--headless", "researcher", "--bottle", "claude", "--prompt", "Do it"]
)
self.assertTrue(self._launch_mock.call_args[1]["assume_yes"])
# -- prompt --------------------------------------------------------
def test_headless_without_prompt_dies(self):
with self.assertRaises(Die):
start_mod.cmd_start(["--headless", "researcher", "--bottle", "claude"])
self._launch_mock.assert_not_called()
def test_headless_prompt_forwarded_to_launch(self):
start_mod.cmd_start(
["--headless", "researcher", "--bottle", "claude",
"--prompt", "Implement issue #42"]
)
self.assertEqual(
"Implement issue #42",
self._launch_mock.call_args[1]["headless_prompt_text"],
)
# -- bottle resolution ---------------------------------------------
def test_explicit_bottles_forwarded_in_order(self):
start_mod.cmd_start(
["--headless", "researcher", "--bottle", "dev", "--bottle", "claude",
"--prompt", "Do it"]
)
self.assertEqual(("dev", "claude"), self._spec().bottle_names)
def test_omitted_bottle_falls_back_to_agent_default(self):
start_mod.cmd_start(["--headless", "implementer", "--prompt", "Do it"])
self.assertEqual(("claude",), self._spec().bottle_names)
def test_no_bottle_and_no_default_dies(self):
manifest = _make_manifest(["researcher"], ["claude"], agent_bottle="")
with patch(
"bot_bottle.cli.start.ManifestIndex.resolve", return_value=manifest
):
with self.assertRaises(Die):
start_mod.cmd_start(
["--headless", "researcher", "--prompt", "Do it"]
)
self._launch_mock.assert_not_called()
# -- agent resolution ----------------------------------------------
def test_missing_agent_name_dies(self):
with self.assertRaises(Die):
start_mod.cmd_start(["--headless"])
self._launch_mock.assert_not_called()
def test_unknown_agent_raises_manifest_error(self):
self._manifest.require_agent.side_effect = ManifestError("agent 'x' not defined")
with self.assertRaises(ManifestError):
start_mod.cmd_start(
["--headless", "x", "--bottle", "claude", "--prompt", "Do it"]
)
self._launch_mock.assert_not_called()
# -- label / color -------------------------------------------------
def test_label_defaults_to_agent_name(self):
start_mod.cmd_start(
["--headless", "researcher", "--bottle", "claude", "--prompt", "Do it"]
)
self.assertEqual("researcher", self._spec().label)
def test_explicit_label_and_color_forwarded(self):
start_mod.cmd_start(
["--headless", "researcher", "--bottle", "claude",
"--label", "nightly", "--color", "green", "--prompt", "Do it"]
)
spec = self._spec()
self.assertEqual("nightly", spec.label)
self.assertEqual("green", spec.color)
def test_label_collision_uniquifies(self):
with patch(
"bot_bottle.cli.start.enumerate_active_agents",
return_value=[_active_agent("researcher")],
):
start_mod.cmd_start(
["--headless", "researcher", "--bottle", "claude", "--prompt", "Do it"]
)
self.assertEqual("researcher-2", self._spec().label)
# -- backend wiring ------------------------------------------------
def test_backend_flag_forwarded(self):
start_mod.cmd_start(
["--headless", "--backend=docker", "researcher", "--bottle", "claude",
"--prompt", "Do it"]
)
self.assertEqual("docker", self._launch_mock.call_args[1]["backend_name"])
class TestPrepareWithPreflight(unittest.TestCase):
"""prepare_with_preflight calls render_preflight with the plan and backend name."""
def test_render_preflight_called(self):
from pathlib import Path
mock_backend = MagicMock()
mock_plan = MagicMock()
mock_backend.prepare.return_value = mock_plan
mock_backend.name = "test-backend"
render = MagicMock()
with patch("bot_bottle.cli.start.get_bottle_backend", return_value=mock_backend), \
patch("bot_bottle.cli.start._identity_from_plan", return_value="id"), \
patch("bot_bottle.cli.start.info"):
start_mod.prepare_with_preflight(
MagicMock(),
stage_dir=Path("/tmp"),
render_preflight=render,
prompt_yes=MagicMock(),
dry_run=True,
)
render.assert_called_once_with(mock_plan, "test-backend")
class TestNoAgentsDefined(unittest.TestCase):
"""cmd_start prints an error and returns 1 when the manifest has no agents."""
def test_no_agents_defined_returns_1(self):
manifest = _make_manifest([], [])
with patch(
"bot_bottle.cli.start.ManifestIndex.resolve", return_value=manifest
), patch("sys.stderr", io.StringIO()) as err:
rc = start_mod.cmd_start([])
self.assertEqual(1, rc)
self.assertIn("no agents defined", err.getvalue())
class TestTextRenderPreflight(unittest.TestCase):
"""_text_render_preflight returns a renderer that prints the backend name."""
def test_backend_name_in_output(self):
render = start_mod._text_render_preflight()
plan = MagicMock()
with patch("bot_bottle.cli.start._manifest_to_yaml", return_value=""), \
patch("sys.stderr", io.StringIO()) as err:
render(plan, "my-backend")
self.assertIn("my-backend", err.getvalue())
if __name__ == "__main__":
unittest.main()
+3 -4
View File
@@ -107,7 +107,7 @@ def _egress_plan(
def _supervise_plan() -> SupervisePlan:
return SupervisePlan(
slug=SLUG,
queue_dir=STATE / "supervise" / "queue",
db_path=STATE / "bot-bottle.db",
internal_network=f"bot-bottle-net-{SLUG}",
)
@@ -392,7 +392,7 @@ class TestSidecarBundleShape(unittest.TestCase):
sc = self._render(supervise=True)["services"]["sidecars"]
env_strings = sc["environment"]
self.assertIn(f"SUPERVISE_BOTTLE_SLUG={SLUG}", env_strings)
self.assertTrue(any(e.startswith("SUPERVISE_QUEUE_DIR=") for e in env_strings))
self.assertIn("SUPERVISE_DB_PATH=/run/supervise/bot-bottle.db", env_strings)
self.assertTrue(any(e.startswith("SUPERVISE_PORT=") for e in env_strings))
def test_volumes_always_includes_egress_ca(self):
@@ -408,8 +408,7 @@ class TestSidecarBundleShape(unittest.TestCase):
self.assertIn("/etc/egress", targets)
self.assertIn("/git-gate-entrypoint.sh", targets)
self.assertIn("/git-gate/creds/upstream-known_hosts", targets)
self.assertTrue(any("supervise/queue" in t or t.startswith("/run/supervise")
for t in targets))
self.assertIn("/run/supervise/bot-bottle.db", targets)
def test_extra_hosts_omitted_for_git_upstreams(self):
sc = self._render(with_git=True)["services"]["sidecars"]
+10 -1
View File
@@ -74,7 +74,7 @@ def _plan(
if supervise:
supervise_plan = SupervisePlan(
slug="demo-abc12",
queue_dir=Path("/tmp/queue"),
db_path=Path("/tmp/bot-bottle.db"),
)
return DockerBottlePlan(
spec=spec,
@@ -343,5 +343,14 @@ class TestClaudeSuperviseMcp(unittest.TestCase):
)
class TestClaudeHeadlessPrompt(unittest.TestCase):
def test_returns_p_flag_and_prompt(self):
self.assertEqual(["-p", "Do the task"], ClaudeAgentProvider().headless_prompt("Do the task"))
def test_preserves_prompt_text_verbatim(self):
text = "Fix issue #42: the widget breaks on empty input"
self.assertEqual(["-p", text], ClaudeAgentProvider().headless_prompt(text))
if __name__ == "__main__":
unittest.main()
+42 -2
View File
@@ -77,7 +77,7 @@ def _plan(
if supervise:
supervise_plan = SupervisePlan(
slug="demo-abc12",
queue_dir=Path("/tmp/queue"),
db_path=Path("/tmp/bot-bottle.db"),
)
return DockerBottlePlan(
spec=spec,
@@ -261,6 +261,36 @@ class TestCodexProvision(unittest.TestCase):
self.assertTrue(any("find" in s and "-delete" in s for s in scripts))
self.assertTrue(any("runuser" in s and "codex login status" in s for s in scripts))
def test_forwarded_credentials_verify_sets_codex_path(self):
with tempfile.TemporaryDirectory(prefix="bb-codex-auth.") as tmp:
state_dir = Path(tmp)
prompt_file = state_dir / "prompt.txt"
prompt_file.write_text("")
with patch(
"bot_bottle.contrib.codex.agent_provider.codex_host_access_token",
return_value="token",
), patch(
"bot_bottle.contrib.codex.agent_provider.write_codex_dummy_auth_file",
):
provision = CodexAgentProvider().provision_plan(
dockerfile="",
state_dir=state_dir,
instance_name="bot-bottle-demo-abc12",
prompt_file=prompt_file,
forward_host_credentials=True,
host_env={},
)
verify_argv = provision.verify[0].argv
self.assertTrue(any(
item.startswith("PATH=/home/node/.local/bin:")
for item in verify_argv
))
self.assertIn(
"/home/node/.codex/packages/standalone/current/bin/codex",
verify_argv,
)
def test_dies_when_dir_creation_fails(self):
provision = AgentProvisionPlan(
template="codex", command="codex",
@@ -301,7 +331,8 @@ class TestCodexSuperviseMcp(unittest.TestCase):
script = bottle.exec.call_args.args[0]
self.assertEqual("node", bottle.exec.call_args.kwargs.get("user"))
self.assertEqual(
f"codex mcp add supervise --url {_URL}",
"/home/node/.codex/packages/standalone/current/bin/codex "
f"mcp add supervise --url {_URL}",
script,
)
@@ -314,5 +345,14 @@ class TestCodexSuperviseMcp(unittest.TestCase):
)
class TestCodexHeadlessPrompt(unittest.TestCase):
def test_returns_prompt_as_positional_arg(self):
self.assertEqual(["Do the task"], CodexAgentProvider().headless_prompt("Do the task"))
def test_preserves_prompt_text_verbatim(self):
text = "Fix issue #42: the widget breaks on empty input"
self.assertEqual([text], CodexAgentProvider().headless_prompt(text))
if __name__ == "__main__":
unittest.main()
+9
View File
@@ -223,5 +223,14 @@ class TestPiDockerfile(unittest.TestCase):
self.assertIn("chmod 1777 /tmp /var/tmp", dockerfile)
class TestPiHeadlessPrompt(unittest.TestCase):
def test_returns_p_flag_and_prompt(self):
self.assertEqual(["-p", "Do the task"], PiAgentProvider().headless_prompt("Do the task"))
def test_preserves_prompt_text_verbatim(self):
text = "Fix issue #42: the widget breaks on empty input"
self.assertEqual(["-p", text], PiAgentProvider().headless_prompt(text))
if __name__ == "__main__":
unittest.main()
@@ -38,6 +38,7 @@ class _Provider(AgentProvider):
def provision_prompt(self, plan, bottle): ... # type: ignore[override]
def provision(self, plan, bottle): ... # type: ignore[override]
def provision_supervise_mcp(self, plan, bottle, supervise_url): ... # type: ignore[override]
def headless_prompt(self, prompt): return [] # type: ignore[override]
_PROVIDER = _Provider()
+3 -3
View File
@@ -19,7 +19,7 @@ from bot_bottle.egress import (
egress_sidecar_env_entries,
egress_token_env_map,
)
from bot_bottle.log import Die
from bot_bottle.errors import MissingEnvVarError
from bot_bottle.manifest import ManifestIndex
from bot_bottle.yaml_subset import parse_yaml_subset
@@ -499,14 +499,14 @@ class TestResolveTokenValues(unittest.TestCase):
self.assertEqual({"EGRESS_TOKEN_0": "the-value"}, out)
def test_missing_token_ref_dies(self):
with self.assertRaises(Die):
with self.assertRaises(MissingEnvVarError):
egress_resolve_token_values(
{"EGRESS_TOKEN_0": "GH_PAT"},
{},
)
def test_empty_token_ref_dies(self):
with self.assertRaises(Die):
with self.assertRaises(MissingEnvVarError):
egress_resolve_token_values(
{"EGRESS_TOKEN_0": "GH_PAT"},
{"GH_PAT": ""},
@@ -47,7 +47,6 @@ def _addon() -> EgressAddon:
a: EgressAddon = EgressAddon.__new__(EgressAddon)
a.config = Config(routes=(), log=LOG_FULL)
a.safe_tokens = set()
a._supervise_queue_dir = ""
a._supervise_slug = ""
a._token_allow_timeout = 300.0
return a
+3 -6
View File
@@ -212,7 +212,6 @@ def _addon(config: Config) -> EgressAddon:
a: EgressAddon = EgressAddon.__new__(EgressAddon)
a.config = config
a.safe_tokens = set()
a._supervise_queue_dir = ""
a._supervise_slug = ""
a._token_allow_timeout = 300.0
a.routes_path = "/nonexistent/routes.yaml"
@@ -386,10 +385,10 @@ def _fake_sv(response_status: str | None) -> types.SimpleNamespace:
def _sha256_hex(_payload: Any) -> str:
return "hash"
def _noop(_a: Any, _b: Any) -> None:
def _noop(*_args: Any) -> None:
return None
def _read_response(_qd: Any, _pid: Any) -> Any:
def _read_response(_slug: Any, _pid: Any) -> Any:
if response_status is None:
raise OSError("not written yet") # forces poll -> timeout
return types.SimpleNamespace(status=response_status)
@@ -409,7 +408,6 @@ def _fake_sv(response_status: str | None) -> types.SimpleNamespace:
class TestSuperviseBranch(unittest.TestCase):
def _supervised_addon(self) -> EgressAddon:
addon = _addon(Config(routes=(Route(host="api.example.com"),)))
addon._supervise_queue_dir = "/tmp/egress-queue"
addon._supervise_slug = "test-bottle"
addon._token_allow_timeout = 0.05
return addon
@@ -632,14 +630,13 @@ class TestRedactSurfaces(unittest.TestCase):
class TestSuperviseWriteFailure(unittest.TestCase):
def test_write_proposal_oserror_blocks(self) -> None:
addon = _addon(Config(routes=(Route(host="api.example.com"),)))
addon._supervise_queue_dir = "/tmp/egress-queue"
addon._supervise_slug = "test-bottle"
addon._token_allow_timeout = 0.05
flow = _Flow(_Request(host="api.example.com", method="POST", body=f"k={_OPENAI_KEY}"))
fake = _fake_sv("approved")
def _raise(_qd: Any, _p: Any) -> None:
def _raise(_p: Any) -> None:
raise OSError("disk full")
fake.write_proposal = _raise
+37 -3
View File
@@ -14,10 +14,12 @@ from bot_bottle.git_gate import (
git_gate_render_access_hook,
git_gate_render_entrypoint,
git_gate_render_hook,
provision_git_gate_dynamic_keys,
revoke_git_gate_provisioned_keys,
_resolve_identity_file,
git_gate_upstreams_for_bottle,
)
from bot_bottle.errors import MissingEnvVarError
from bot_bottle.manifest import ManifestIndex
from tests.fixtures import fixture_minimal, fixture_with_git
@@ -209,12 +211,23 @@ class TestHookRender(unittest.TestCase):
# the suppressed findings for human approval.
self.assertIn("--ignore-gitleaks-allow", hook)
self.assertIn("--report-format=json", hook)
self.assertIn('"tool": "gitleaks-allow"', hook)
self.assertIn("SUPERVISE_QUEUE_DIR", hook)
self.assertIn("tool=_sv.TOOL_GITLEAKS_ALLOW", hook)
self.assertIn("_sv.write_proposal", hook)
self.assertIn("_sv.read_response", hook)
self.assertIn("SUPERVISE_BOTTLE_SLUG", hook)
self.assertIn("supervisor approved # gitleaks:allow", hook)
self.assertIn("supervisor rejected # gitleaks:allow", hook)
def test_inline_gitleaks_allow_python_imports_work_in_sidecar_layout(self):
hook = git_gate_render_hook()
# The sidecar image copies supervise.py flat under /app, while
# host-side tests import it through the bot_bottle package.
# Hooks execute from the bare repo directory, so the embedded
# Python must include /app and support both import layouts.
self.assertIn('PYTHONPATH="/app${PYTHONPATH:+:$PYTHONPATH}"', hook)
self.assertIn("import supervise as _sv", hook)
self.assertIn("from bot_bottle import supervise as _sv", hook)
def test_inline_gitleaks_allow_fails_closed_without_supervisor(self):
hook = git_gate_render_hook()
self.assertIn(
@@ -371,6 +384,27 @@ class TestDynamicKeyProvisioning(unittest.TestCase):
self.assertEqual("/tmp/provisioned-key", _resolve_identity_file(entry, "demo", self.stage))
mock_provision.assert_called_once()
def test_prepare_defers_gitea_key_provisioning(self):
bottle = self._gitea_manifest().bottles["dev"]
with patch("bot_bottle.git_gate_provision._provision_dynamic_key") as mock_provision:
plan = _StubGate().prepare(bottle, "demo", self.stage)
mock_provision.assert_not_called()
self.assertEqual("", plan.upstreams[0].identity_file)
def test_launch_time_helper_provisions_gitea_keys(self):
bottle = self._gitea_manifest().bottles["dev"]
plan = _StubGate().prepare(bottle, "demo", self.stage)
with patch(
"bot_bottle.git_gate_provision._provision_dynamic_key",
return_value="/tmp/provisioned-key",
) as mock_provision:
updated = provision_git_gate_dynamic_keys(bottle, plan, self.stage)
mock_provision.assert_called_once_with(bottle.git[0], "demo", self.stage)
self.assertEqual("/tmp/provisioned-key", updated.upstreams[0].identity_file)
def test_revoke_skips_non_gitea_and_missing_id_file(self):
revoke_git_gate_provisioned_keys(fixture_with_git().bottles["dev"], self.stage)
@@ -388,7 +422,7 @@ class TestDynamicKeyProvisioning(unittest.TestCase):
def test_revoke_missing_token_raises(self):
bottle = self._gitea_manifest().bottles["dev"]
(self.stage / "repo-deploy-key-id").write_text("123\n")
with patch.dict("os.environ", {}, clear=True), self.assertRaises(RuntimeError) as cm:
with patch.dict("os.environ", {}, clear=True), self.assertRaises(MissingEnvVarError) as cm:
revoke_git_gate_provisioned_keys(bottle, self.stage)
self.assertIn("env var is not set", str(cm.exception))
+3 -2
View File
@@ -13,6 +13,7 @@ from pathlib import Path
from typing import Any, cast
from unittest.mock import patch
from bot_bottle.errors import MissingEnvVarError
from bot_bottle.git_gate import (
_gitconfig_validate_value,
_provision_dynamic_key,
@@ -120,7 +121,7 @@ class TestProvisionDynamicKey(unittest.TestCase):
patch.dict("os.environ", {}, clear=False):
import os
os.environ.pop("GITEA_TOK", None)
with self.assertRaises(RuntimeError):
with self.assertRaises(MissingEnvVarError):
_provision_dynamic_key(_gitea_entry(), "s", Path(d))
@@ -166,7 +167,7 @@ class TestRevokeProvisionedKeys(unittest.TestCase):
import os
os.environ.pop("GITEA_TOK", None)
(Path(d) / "repo-deploy-key-id").write_text("kid123")
with self.assertRaises(RuntimeError):
with self.assertRaises(MissingEnvVarError):
revoke_git_gate_provisioned_keys(_bottle(_gitea_entry()), Path(d))
+13
View File
@@ -30,6 +30,19 @@ class TestMacosContainerBottle(unittest.TestCase):
argv,
)
def test_agent_argv_accepts_absolute_provider_command(self):
command = "/home/node/.codex/packages/standalone/current/bin/codex"
bottle = MacosContainerBottle(
"bot-bottle-dev-abc",
lambda: None,
None,
agent_command=command,
)
with patch.dict(bottle_mod.os.environ, {}, clear=True):
argv = bottle.agent_argv(["run"])
self.assertIn(command, argv)
self.assertNotIn("codex", argv)
def test_agent_argv_includes_workdir(self):
bottle = MacosContainerBottle(
"bot-bottle-dev-abc",
+4 -2
View File
@@ -71,7 +71,9 @@ def _plan(
else:
git_gate_plan = SimpleNamespace(upstreams=())
supervise_plan = (
SimpleNamespace(queue_dir=Path("/state/supervise/queue"))
SimpleNamespace(
db_path=Path("/state/bot-bottle.db"),
)
if supervise else None
)
agent_provision = SimpleNamespace(
@@ -137,7 +139,7 @@ class TestMacosContainerLaunchArgv(unittest.TestCase):
argv,
)
self.assertIn(
"type=bind,source=/state/supervise/queue,target=/run/supervise/queue",
"type=bind,source=/state,target=/run/supervise",
argv,
)
+42 -1
View File
@@ -9,7 +9,11 @@ from __future__ import annotations
import unittest
from bot_bottle.manifest import ManifestBottle, ManifestIndex
from unittest.mock import patch
from bot_bottle.env import resolve_env
from bot_bottle.errors import MissingEnvVarError
from bot_bottle.manifest import Manifest, ManifestBottle, ManifestIndex
from bot_bottle.manifest_agent import (
ManifestAgent,
ManifestAgentProvider,
@@ -165,6 +169,22 @@ class TestAgentValidation(unittest.TestCase):
with self.assertRaises(ManifestError):
ManifestAgent.from_dict("a", {"skills": [5]}, set())
def test_skill_name_rejects_shell_metacharacters(self) -> None:
# Skill names become host/guest path segments interpolated into
# provisioning shell commands; anything outside kebab-case is
# rejected at load so it can never reach a `bottle.exec` string.
for bad in ("foo; rm -rf /", "../escape", "foo bar", "Foo", "-leading"):
with self.assertRaises(ManifestError):
ManifestAgent.from_dict("a", {"skills": [bad]}, set())
def test_skill_name_accepts_kebab_case(self) -> None:
agent = ManifestAgent.from_dict(
"a", {"skills": ["init-entry", "quality-eval", "skill0"]}, set()
)
self.assertEqual(
agent.skills, ("init-entry", "quality-eval", "skill0")
)
def test_prompt_not_string(self) -> None:
with self.assertRaises(ManifestError):
ManifestAgent.from_dict("a", {"prompt": 5}, set())
@@ -222,5 +242,26 @@ class TestEagerIndexLookups(unittest.TestCase):
self.assertIsNone(m.git_identity_summary())
class TestResolveEnv(unittest.TestCase):
"""resolve_env raises MissingEnvVarError when an interpolated entry's
host var is unset."""
def _manifest(self, env_dict: dict[str, str]) -> Manifest:
idx = ManifestIndex.from_json_obj({
"bottles": {"dev": {"env": env_dict}},
"agents": {"demo": {"bottle": "dev"}},
})
return idx.load_for_agent("demo")
def test_missing_interpolated_host_var_raises(self):
manifest = self._manifest({"API_KEY": "${HOST_API_KEY}"})
with patch.dict("os.environ", {}, clear=False):
import os
os.environ.pop("HOST_API_KEY", None)
with self.assertRaises(MissingEnvVarError) as cm:
resolve_env(manifest)
self.assertIn("HOST_API_KEY", str(cm.exception))
if __name__ == "__main__":
unittest.main()
+15 -2
View File
@@ -79,6 +79,11 @@ class TestClaudeArgvWrapped(unittest.TestCase):
"--",
"runuser", "-u", "node", "--",
"env", "HOME=/home/node", "USER=node",
(
"PATH=/home/node/.local/bin:"
"/home/node/.codex/packages/standalone/current/bin:"
"/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
),
"claude",
],
argv,
@@ -127,16 +132,24 @@ class TestClaudeArgvWrapped(unittest.TestCase):
self.assertIn("HTTPS_PROXY=http://127.0.0.1:1234", argv)
self.assertIn("NO_PROXY=localhost", argv)
def test_guest_env_path_overrides_default_path(self):
argv = _unwrap(_bottle(None, PATH="/custom/bin").agent_argv([]))
self.assertIn("PATH=/custom/bin", argv)
self.assertFalse(any(
item.startswith("PATH=/home/node/.local/bin")
for item in argv
))
def test_runuser_switch_precedes_claude(self):
# The dashboard's `_build_resume_argv_with_fallback` finds
# the `claude` token to split exec-framing from the claude
# tail. `runuser -u node --` must sit on the prefix side so
# the shell wrap inherits the UID switch.
argv = _bottle().agent_argv([])
agent_idx = argv.index("claude")
runuser_idx = argv.index("runuser")
self.assertEqual(
["runuser", "-u", "node", "--", "env"],
argv[agent_idx - 7:agent_idx - 2],
argv[runuser_idx:runuser_idx + 5],
)
def test_pi_provider_appends_system_prompt_without_print_mode(self):
+5 -1
View File
@@ -94,6 +94,10 @@ class TestEnsureSmolmachine(unittest.TestCase):
calls.append(name)
return _f
def save_and_record(image_ref: str, path: str) -> None:
Path(path).touch()
calls.append("save")
with patch.object(
_launch_mod.docker_mod, "build_image",
side_effect=record("build"),
@@ -102,7 +106,7 @@ class TestEnsureSmolmachine(unittest.TestCase):
return_value=f"sha256:{digest}fffffffffffffffff",
), patch.object(
_launch_mod.docker_mod, "save",
side_effect=record("save"),
side_effect=save_and_record,
) as save, patch.object(
_launch_mod, "ephemeral_registry",
return_value=_Reg(),
+117 -14
View File
@@ -8,6 +8,7 @@ inspecting running bundle containers' port bindings."""
from __future__ import annotations
import json
import os
import sqlite3
import subprocess
import tempfile
@@ -112,9 +113,16 @@ class TestEnsurePool(unittest.TestCase):
class TestAllocate(unittest.TestCase):
def test_returns_loopback_on_linux(self):
with patch.object(loopback_alias, "_is_macos", return_value=False):
self.assertEqual("127.0.0.1", loopback_alias.allocate("demo"))
def test_per_bottle_alias_on_linux(self):
# Linux gets the same per-bottle scoping as macOS (127/8 is
# already loopback, so no ifconfig is needed). A fresh host
# with no running bundles allocates the first pool entry.
with tempfile.TemporaryDirectory() as tmp:
lock_path = Path(tmp) / "smolmachines.lock"
with patch.object(loopback_alias, "_is_macos", return_value=False), \
patch.object(loopback_alias, "_ALLOC_LOCK_PATH", lock_path), \
patch.object(loopback_alias, "_aliases_in_use", return_value=set()):
self.assertEqual("127.0.0.16", loopback_alias.allocate("demo"))
def test_picks_lowest_unused_on_macos(self):
# No bundles running -> first pool entry.
@@ -166,12 +174,25 @@ class TestAllocateLock(unittest.TestCase):
self.assertIn(fcntl_mod.LOCK_EX, flock_calls)
def test_no_lock_on_linux(self):
# Linux early-returns before touching the lock file.
def test_acquires_exclusive_lock_on_linux(self):
# Linux allocates per-bottle too, so it must take the same
# lock to serialise concurrent launches.
import fcntl as fcntl_mod
flock_calls: list[int] = []
def record_flock(fd, op): # type: ignore
flock_calls.append(op)
with tempfile.TemporaryDirectory() as tmp:
lock_path = Path(tmp) / "smolmachines.lock"
with patch.object(loopback_alias, "_is_macos", return_value=False), \
patch.object(loopback_alias.fcntl, "flock") as flock:
patch.object(loopback_alias, "_ALLOC_LOCK_PATH", lock_path), \
patch.object(loopback_alias, "_aliases_in_use", return_value=set()), \
patch.object(loopback_alias.fcntl, "flock",
side_effect=record_flock):
loopback_alias.allocate("demo")
flock.assert_not_called()
self.assertIn(fcntl_mod.LOCK_EX, flock_calls)
def test_sequential_allocations_with_shared_lock_are_serialised(self):
# Two sequential calls share the same lock file. The second
@@ -241,10 +262,12 @@ class TestAliasInUseDetection(unittest.TestCase):
class TestForceAllowlist(unittest.TestCase):
"""Smolvm 0.8.0 silently drops `--allow-cidr` with `--from`,
so `force_allowlist` opens the state DB directly and sets
the row's `allowed_cidrs` field. Round-trip tests against a
real SQLite DB to lock down the BLOB encoding."""
"""Smolvm 0.8.0 silently drops `--allow-cidr` with `--from`, so
`force_allowlist` opens the state DB directly and sets the row's
`allowed_cidrs` field on both macOS and Linux. It is
fail-closed: it dies rather than launching a VM whose allowlist
it can't confirm. Round-trip tests against a real SQLite DB to
lock down the BLOB encoding."""
def setUp(self):
self._tmp = tempfile.TemporaryDirectory(prefix="smolvm-db.")
@@ -290,17 +313,67 @@ class TestForceAllowlist(unittest.TestCase):
self.assertEqual(4, cfg["cpus"])
self.assertTrue(cfg["network"])
def test_noop_on_linux(self):
def test_patches_on_linux_too(self):
# force_allowlist no longer no-ops on Linux — the TSI
# allowlist must be enforced there as well.
with patch.object(loopback_alias, "_is_macos", return_value=False), \
patch.object(loopback_alias, "_SMOLVM_DB_PATH", self.db):
loopback_alias.force_allowlist("demo-vm", ["127.0.0.16/32"])
# DB row should be untouched.
con = sqlite3.connect(str(self.db))
cfg = json.loads(con.execute(
"SELECT data FROM vms WHERE name='demo-vm'",
).fetchone()[0])
con.close()
self.assertIsNone(cfg["allowed_cidrs"])
self.assertEqual(["127.0.0.16/32"], cfg["allowed_cidrs"])
def test_skips_write_when_already_matching(self):
# A newer smolvm that honors --allow-cidr at create leaves the
# row already correct; force_allowlist must not rewrite it. We
# detect a no-write by comparing the raw BLOB byte-for-byte
# (a rewrite re-serialises the JSON, changing key order/bytes
# is not guaranteed, but mtime/identity isn't observable — so
# we assert the stored bytes are exactly what we pre-seeded).
seeded = json.dumps({
"name": "demo-vm", "cpus": 4, "mem": 8192,
"network": True, "allowed_cidrs": ["127.0.0.16/32"],
}).encode()
con = sqlite3.connect(str(self.db))
con.execute(
"UPDATE vms SET data=? WHERE name='demo-vm'",
(sqlite3.Binary(seeded),),
)
con.commit()
con.close()
with patch.object(loopback_alias, "_is_macos", return_value=True), \
patch.object(loopback_alias, "_SMOLVM_DB_PATH", self.db):
loopback_alias.force_allowlist("demo-vm", ["127.0.0.16/32"])
con = sqlite3.connect(str(self.db))
stored = con.execute(
"SELECT data FROM vms WHERE name='demo-vm'").fetchone()[0]
con.close()
self.assertEqual(seeded, bytes(stored))
def test_dies_when_patch_does_not_take(self):
# If the persisted allowlist still doesn't match after the
# patch (e.g. wrong schema / smolvm stores it elsewhere),
# force_allowlist must fail closed rather than boot the VM.
original = loopback_alias._read_machine_cfg
def stale_cfg(con: sqlite3.Connection, name: str) -> dict[str, object]:
# Always report the un-patched row so the post-write
# verification never sees the requested cidrs.
cfg = original(con, name)
cfg["allowed_cidrs"] = None
return cfg
with patch.object(loopback_alias, "_is_macos", return_value=True), \
patch.object(loopback_alias, "_SMOLVM_DB_PATH", self.db), \
patch.object(loopback_alias, "_read_machine_cfg", side_effect=stale_cfg), \
patch.object(loopback_alias, "die", side_effect=SystemExit("die")):
with self.assertRaises(SystemExit):
loopback_alias.force_allowlist("demo-vm", ["127.0.0.16/32"])
def test_dies_on_missing_db(self):
with patch.object(loopback_alias, "_is_macos", return_value=True), \
@@ -323,5 +396,35 @@ class TestForceAllowlist(unittest.TestCase):
loopback_alias.force_allowlist("not-in-db", ["127.0.0.16/32"])
class TestSmolvmDbPath(unittest.TestCase):
"""The smolvm state-DB path is platform-derived: Application
Support on macOS, XDG data dir on Linux."""
def test_macos_path(self):
with patch.object(loopback_alias.platform, "system", return_value="Darwin"):
p = loopback_alias._smolvm_db_path()
self.assertEqual(
("Library", "Application Support", "smolvm", "server", "smolvm.db"),
p.parts[-5:],
)
def test_linux_default_xdg_path(self):
env = {k: v for k, v in os.environ.items() if k != "XDG_DATA_HOME"}
with patch.object(loopback_alias.platform, "system", return_value="Linux"), \
patch.dict(loopback_alias.os.environ, env, clear=True):
p = loopback_alias._smolvm_db_path()
self.assertEqual(
(".local", "share", "smolvm", "server", "smolvm.db"),
p.parts[-5:],
)
def test_linux_respects_xdg_data_home(self):
with patch.object(loopback_alias.platform, "system", return_value="Linux"), \
patch.dict(loopback_alias.os.environ,
{"XDG_DATA_HOME": "/custom/data"}, clear=False):
p = loopback_alias._smolvm_db_path()
self.assertEqual(Path("/custom/data/smolvm/server/smolvm.db"), p)
if __name__ == "__main__":
unittest.main()
+61 -3
View File
@@ -49,6 +49,7 @@ class _Provider(AgentProvider):
def provision_prompt(self, plan, bottle): ... # type: ignore[override]
def provision(self, plan, bottle): ... # type: ignore[override]
def provision_supervise_mcp(self, plan, bottle, supervise_url): ... # type: ignore[override]
def headless_prompt(self, prompt): return [] # type: ignore[override]
_PROVIDER = _Provider()
@@ -129,7 +130,7 @@ def _plan(
if supervise:
supervise_plan = SupervisePlan(
slug="demo-abc12",
queue_dir=Path("/tmp/queue"),
db_path=Path("/tmp/bot-bottle.db"),
)
return SmolmachinesBottlePlan(
spec=spec,
@@ -275,8 +276,8 @@ class TestProvisionCA(unittest.TestCase):
str(self.egress_ca),
AGENT_CA_PATH,
)
bottle.exec.assert_called_once()
script = bottle.exec.call_args.args[0]
self.assertEqual(2, bottle.exec.call_count)
script = bottle.exec.call_args_list[1].args[0]
self.assertIn("chmod 644", script)
self.assertIn("update-ca-certificates", script)
self.assertEqual("root", bottle.exec.call_args.kwargs.get("user"))
@@ -421,6 +422,14 @@ class TestBundleLaunchSpec(unittest.TestCase):
spec.environment,
)
def test_supervise_adds_daemon_volume_and_env(self):
from bot_bottle.supervise import DB_PATH_IN_CONTAINER
plan = _plan(supervise=True)
spec = _bundle_launch_spec(plan, "net", "127.0.0.16")
self.assertIn("supervise", spec.daemons_csv)
self.assertIn(f"SUPERVISE_DB_PATH={DB_PATH_IN_CONTAINER}", spec.environment)
self.assertIn(("/tmp/bot-bottle.db", DB_PATH_IN_CONTAINER, False), spec.volumes)
def test_canary_env_visible_to_smolvm_guest(self):
plan = _plan(canary=True)
with patch.object(
@@ -497,5 +506,54 @@ class TestProvisionGitUser(unittest.TestCase):
self.assertIn("bot@example.com", calls[0][0])
class TestProxyHost(unittest.TestCase):
"""_proxy_host returns the bridge gateway on Linux and the loopback
alias on other platforms."""
def test_linux_returns_bundle_gateway(self):
plan = _plan()
with patch("bot_bottle.backend.smolmachines.launch.platform.system",
return_value="Linux"):
result = _launch._proxy_host(plan, "127.0.0.16")
self.assertEqual(plan.bundle_gateway, result)
def test_non_linux_returns_loopback(self):
plan = _plan()
with patch("bot_bottle.backend.smolmachines.launch.platform.system",
return_value="Darwin"):
result = _launch._proxy_host(plan, "127.0.0.16")
self.assertEqual("127.0.0.16", result)
class TestDiscoverUrls(unittest.TestCase):
"""_discover_urls stamps git-gate host + supervise URL into the plan."""
def test_git_gate_host_set_when_upstreams_present(self):
plan = _plan()
plan = replace(
plan,
git_gate_plan=replace(
plan.git_gate_plan,
upstreams=(GitGateUpstream(
name="bot-bottle",
upstream_url="ssh://git@host/repo.git",
upstream_host="host",
upstream_port="22",
identity_file="/tmp/key",
known_host_key="",
),),
),
)
with patch.object(_launch._bundle, "bundle_host_port", return_value="9420"):
stamped = _launch._discover_urls(plan, "127.0.0.16")
self.assertEqual("127.0.0.16:9420", stamped.agent_git_gate_host)
def test_supervise_url_set_when_supervise_present(self):
plan = _plan(supervise=True)
with patch.object(_launch._bundle, "bundle_host_port", return_value="55556"):
stamped = _launch._discover_urls(plan, "127.0.0.16")
self.assertEqual("http://127.0.0.16:55556/", stamped.agent_supervise_url)
if __name__ == "__main__":
unittest.main()
+6 -5
View File
@@ -79,7 +79,7 @@ class TestArgvShapes(unittest.TestCase):
with self._patch_run() as m:
machine_create("agent-xyz")
self.assertEqual(
["smolvm", "machine", "create", "agent-xyz"],
["smolvm", "machine", "create", "--name", "agent-xyz"],
m.call_args.args[0],
)
@@ -103,7 +103,8 @@ class TestArgvShapes(unittest.TestCase):
self.assertIn("192.168.50.2/32", argv)
self.assertIn("-e", argv)
self.assertIn("HTTPS_PROXY=http://192.168.50.2:8888", argv)
self.assertEqual("agent-xyz", argv[-1])
self.assertIn("--name", argv)
self.assertIn("agent-xyz", argv)
def test_machine_create_omits_net_when_no_allow_cidrs(self):
with self._patch_run() as m:
@@ -127,13 +128,13 @@ class TestArgvShapes(unittest.TestCase):
m.call_args.args[0],
)
def test_machine_delete_uses_positional_name_and_force(self):
# delete NAME is positional; -f required so no interactive
def test_machine_delete_uses_name_flag_and_force(self):
# delete uses --name flag; -f required so no interactive
# confirmation blocks teardown.
with self._patch_run() as m:
machine_delete("agent-xyz")
self.assertEqual(
["smolvm", "machine", "delete", "-f", "agent-xyz"],
["smolvm", "machine", "delete", "--name", "agent-xyz", "-f"],
m.call_args.args[0],
)
+63
View File
@@ -56,9 +56,14 @@ class TestBundleSubnet(unittest.TestCase):
class TestPreflight(unittest.TestCase):
def test_smolvm_present_returns_none(self):
# Pin macOS so the Linux KVM gate doesn't fire on a CI runner
# (ubuntu, no /dev/kvm) — this test isolates the PATH check.
with patch(
"bot_bottle.backend.smolmachines.util.shutil.which",
return_value="/usr/local/bin/smolvm",
), patch(
"bot_bottle.backend.smolmachines.util.platform.system",
return_value="Darwin",
):
self.assertIsNone(smolmachines_preflight())
@@ -88,5 +93,63 @@ class TestPreflight(unittest.TestCase):
self.assertIn("BOT_BOTTLE_BACKEND=docker", msg)
class TestKvmPreflight(unittest.TestCase):
"""Linux-only KVM gate: smolvm needs /dev/kvm present and
accessible. macOS skips this entirely (Hypervisor.framework)."""
def _run(self, *, system: str, exists: bool, access: bool) -> None:
with patch(
"bot_bottle.backend.smolmachines.util.shutil.which",
return_value="/usr/bin/smolvm",
), patch(
"bot_bottle.backend.smolmachines.util.platform.system",
return_value=system,
), patch(
"bot_bottle.backend.smolmachines.util.os.path.exists",
return_value=exists,
), patch(
"bot_bottle.backend.smolmachines.util.os.access",
return_value=access,
):
return smolmachines_preflight()
def test_macos_skips_kvm_check(self):
# Even with /dev/kvm absent, macOS must not run the gate.
self.assertIsNone(self._run(system="Darwin", exists=False, access=False))
def test_linux_ok_returns_none(self):
self.assertIsNone(self._run(system="Linux", exists=True, access=True))
def test_linux_missing_device_dies(self):
with self.assertRaises(SystemExit):
self._run(system="Linux", exists=False, access=False)
def test_linux_no_access_dies(self):
with self.assertRaises(SystemExit):
self._run(system="Linux", exists=True, access=False)
def test_linux_missing_device_message(self):
import io
import sys
captured = io.StringIO()
with patch.object(sys, "stderr", captured):
with self.assertRaises(SystemExit):
self._run(system="Linux", exists=False, access=False)
msg = captured.getvalue()
self.assertIn("/dev/kvm", msg)
self.assertIn("kvm-intel", msg)
def test_linux_no_access_message(self):
import io
import sys
captured = io.StringIO()
with patch.object(sys, "stderr", captured):
with self.assertRaises(SystemExit):
self._run(system="Linux", exists=True, access=False)
msg = captured.getvalue()
self.assertIn("kvm", msg)
self.assertIn("group", msg)
if __name__ == "__main__":
unittest.main()
+78 -40
View File
@@ -1,6 +1,5 @@
"""Unit: supervise queue + audit log + diff helpers (PRD 0013)."""
import json
import tempfile
import threading
import time
@@ -9,6 +8,9 @@ from datetime import datetime, timezone
from pathlib import Path
from bot_bottle import supervise
from bot_bottle import supervise_types as sv_types
from bot_bottle.audit_store import AuditStore
from bot_bottle.queue_store import QueueStore
from bot_bottle.supervise import (
AuditEntry,
Proposal,
@@ -19,7 +21,7 @@ from bot_bottle.supervise import (
TOOL_EGRESS_ALLOW,
TOOL_GITLEAKS_ALLOW,
archive_proposal,
audit_log_path,
host_db_path,
list_pending_proposals,
read_audit_entries,
read_proposal,
@@ -112,32 +114,52 @@ class TestResponseRoundtrip(unittest.TestCase):
class TestQueueIO(unittest.TestCase):
def setUp(self):
self._tmp = tempfile.TemporaryDirectory(prefix="bot-bottle-supervise-test.")
self.queue_dir = Path(self._tmp.name)
self._home_patch = self._patch_home(Path(self._tmp.name))
self.slug = "dev"
QueueStore("").migrate()
def tearDown(self):
self._home_patch()
self._tmp.cleanup()
def _patch_home(self, fake_home: Path):
original_sv = supervise.bot_bottle_root
original_svt = sv_types.bot_bottle_root
def fake_root() -> Path:
return fake_home / ".bot-bottle"
supervise.bot_bottle_root = fake_root # type: ignore[assignment]
sv_types.bot_bottle_root = fake_root # type: ignore[assignment]
def restore() -> None:
supervise.bot_bottle_root = original_sv # type: ignore[assignment]
sv_types.bot_bottle_root = original_svt # type: ignore[assignment]
return restore
def test_write_and_read_proposal(self):
p = _proposal()
path = write_proposal(self.queue_dir, p)
path = write_proposal(p)
self.assertTrue(path.exists())
self.assertEqual(host_db_path(), path)
self.assertEqual(0o600, path.stat().st_mode & 0o777)
loaded = read_proposal(self.queue_dir, p.id)
loaded = read_proposal(self.slug, p.id)
self.assertEqual(p, loaded)
def test_list_pending_excludes_responded(self):
a = _proposal(justification="first")
b = _proposal(justification="second")
write_proposal(self.queue_dir, a)
write_proposal(self.queue_dir, b)
write_response(self.queue_dir, Response(
write_proposal(a)
write_proposal(b)
write_response(self.slug, Response(
proposal_id=a.id, status=STATUS_APPROVED, notes="",
))
pending = list_pending_proposals(self.queue_dir)
pending = list_pending_proposals(self.slug)
self.assertEqual([b.id], [p.id for p in pending])
def test_list_pending_returns_empty_for_missing_dir(self):
self.assertEqual([], list_pending_proposals(self.queue_dir / "nope"))
def test_list_pending_returns_empty_for_missing_slug(self):
self.assertEqual([], list_pending_proposals("nope"))
def test_list_pending_sorted_by_arrival(self):
# Fabricate two with explicit timestamps.
@@ -154,30 +176,30 @@ class TestQueueIO(unittest.TestCase):
now=datetime(2026, 5, 25, 14, 0, 0, tzinfo=timezone.utc),
)
# Write in reverse order.
write_proposal(self.queue_dir, b)
write_proposal(self.queue_dir, a)
ordered = list_pending_proposals(self.queue_dir)
write_proposal(b)
write_proposal(a)
ordered = list_pending_proposals(self.slug)
self.assertEqual([a.id, b.id], [p.id for p in ordered])
def test_write_and_read_response(self):
r = Response(proposal_id="xyz", status=STATUS_REJECTED, notes="no")
write_response(self.queue_dir, r)
self.assertEqual(r, read_response(self.queue_dir, "xyz"))
write_response(self.slug, r)
self.assertEqual(r, read_response(self.slug, "xyz"))
def test_wait_for_response_returns_when_file_appears(self):
p = _proposal()
write_proposal(self.queue_dir, p)
write_proposal(p)
def write_after_delay():
time.sleep(0.05)
write_response(self.queue_dir, Response(
write_response(self.slug, Response(
proposal_id=p.id, status=STATUS_APPROVED, notes="ok",
))
t = threading.Thread(target=write_after_delay)
t.start()
try:
r = wait_for_response(self.queue_dir, p.id, poll_interval=0.01)
r = wait_for_response(self.slug, p.id, poll_interval=0.01)
finally:
t.join()
self.assertEqual(STATUS_APPROVED, r.status)
@@ -187,44 +209,51 @@ class TestQueueIO(unittest.TestCase):
deadline = time.monotonic() + 0.05
with self.assertRaises(TimeoutError):
wait_for_response(
self.queue_dir, "never",
self.slug, "never",
poll_interval=0.01, deadline=deadline,
)
def test_archive_proposal_moves_both_files(self):
def test_archive_proposal_hides_rows(self):
p = _proposal()
write_proposal(self.queue_dir, p)
write_response(self.queue_dir, Response(
write_proposal(p)
write_response(self.slug, Response(
proposal_id=p.id, status=STATUS_APPROVED, notes="",
))
archive_proposal(self.queue_dir, p.id)
self.assertFalse((self.queue_dir / f"{p.id}.proposal.json").exists())
self.assertFalse((self.queue_dir / f"{p.id}.response.json").exists())
self.assertTrue((self.queue_dir / "processed" / f"{p.id}.proposal.json").exists())
self.assertTrue((self.queue_dir / "processed" / f"{p.id}.response.json").exists())
archive_proposal(self.slug, p.id)
self.assertEqual([], list_pending_proposals(self.slug))
with self.assertRaises(FileNotFoundError):
read_response(self.slug, p.id)
def test_archive_is_idempotent_on_missing_files(self):
# Should not raise.
archive_proposal(self.queue_dir, "nope")
archive_proposal(self.slug, "nope")
class TestAuditLog(unittest.TestCase):
def setUp(self):
self._tmp = tempfile.TemporaryDirectory(prefix="bot-bottle-supervise-audit.")
self._home_patch = self._patch_home(Path(self._tmp.name))
AuditStore().migrate()
def tearDown(self):
self._home_patch()
self._tmp.cleanup()
def _patch_home(self, fake_home: Path):
original = supervise.bot_bottle_root
original_sv = supervise.bot_bottle_root
original_svt = sv_types.bot_bottle_root
def fake_root() -> Path:
return fake_home / ".bot-bottle"
supervise.bot_bottle_root = fake_root # type: ignore[assignment]
return lambda: setattr(supervise, "bot_bottle_root", original)
sv_types.bot_bottle_root = fake_root # type: ignore[assignment]
def restore() -> None:
supervise.bot_bottle_root = original_sv # type: ignore[assignment]
sv_types.bot_bottle_root = original_svt # type: ignore[assignment]
return restore
def test_write_then_read_single_entry(self):
e = AuditEntry(
@@ -237,6 +266,7 @@ class TestAuditLog(unittest.TestCase):
diff="--- before\n+++ after\n",
)
path = write_audit_entry(e)
self.assertEqual(host_db_path(), path)
self.assertEqual(0o600, path.stat().st_mode & 0o777)
loaded = read_audit_entries("cred-proxy", "dev")
self.assertEqual([e], loaded)
@@ -252,12 +282,13 @@ class TestAuditLog(unittest.TestCase):
justification="",
diff="",
))
path = audit_log_path("egress", "dev")
with path.open() as f:
lines = [line for line in f if line.strip()]
self.assertEqual(3, len(lines))
for line in lines:
self.assertTrue(json.loads(line)) # each line is valid JSON
entries = read_audit_entries("egress", "dev")
self.assertEqual(3, len(entries))
self.assertEqual(
["2026-05-25T12:00:00+00:00", "2026-05-25T12:00:01+00:00",
"2026-05-25T12:00:02+00:00"],
[entry.timestamp for entry in entries],
)
def test_separate_logs_per_component_slug(self):
write_audit_entry(AuditEntry(
@@ -369,17 +400,24 @@ class TestSupervisePrepare(unittest.TestCase):
self._tmp.cleanup()
def _patch_home(self, fake_home: Path):
original = supervise.bot_bottle_root
original_sv = supervise.bot_bottle_root
original_svt = sv_types.bot_bottle_root
def fake_root() -> Path:
return fake_home / ".bot-bottle"
supervise.bot_bottle_root = fake_root # type: ignore[assignment]
return lambda: setattr(supervise, "bot_bottle_root", original)
sv_types.bot_bottle_root = fake_root # type: ignore[assignment]
def restore() -> None:
supervise.bot_bottle_root = original_sv # type: ignore[assignment]
sv_types.bot_bottle_root = original_svt # type: ignore[assignment]
return restore
def test_prepare_creates_queue(self):
plan = _StubSupervise().prepare("dev", self.stage_dir)
self.assertTrue(plan.queue_dir.is_dir())
self.assertTrue(plan.db_path.is_file())
self.assertEqual("dev", plan.slug)
self.assertEqual("", plan.internal_network)
+29 -29
View File
@@ -12,7 +12,10 @@ from pathlib import Path
from unittest.mock import patch
from bot_bottle import supervise
from bot_bottle import supervise_types as sv_types
from bot_bottle.audit_store import AuditStore
from bot_bottle.cli import supervise as supervise_cli
from bot_bottle.queue_store import QueueStore
from bot_bottle.supervise import (
Proposal,
STATUS_APPROVED,
@@ -52,13 +55,22 @@ class _FakeHomeMixin:
def _setup_fake_home(self):
self._tmp = tempfile.TemporaryDirectory(prefix="supervise-test.")
original = supervise.bot_bottle_root
original_sv = supervise.bot_bottle_root
original_svt = sv_types.bot_bottle_root
def fake_root() -> Path:
return Path(self._tmp.name) / ".bot-bottle"
supervise.bot_bottle_root = fake_root # type: ignore[assignment]
self._restore_home = lambda: setattr(supervise, "bot_bottle_root", original)
sv_types.bot_bottle_root = fake_root # type: ignore[assignment]
def restore() -> None:
supervise.bot_bottle_root = original_sv # type: ignore[assignment]
sv_types.bot_bottle_root = original_svt # type: ignore[assignment]
self._restore_home = restore
QueueStore("").migrate()
AuditStore().migrate()
def _teardown_fake_home(self):
self._restore_home()
@@ -77,9 +89,7 @@ class TestDiscoverPending(_FakeHomeMixin, unittest.TestCase):
def test_walks_all_slug_subdirs(self):
for slug in ("dev", "api"):
qdir = supervise.queue_dir_for_slug(slug)
qdir.mkdir(parents=True)
supervise.write_proposal(qdir, _proposal(slug=slug))
supervise.write_proposal(_proposal(slug=slug))
pending = supervise_cli.discover_pending()
self.assertEqual({"dev", "api"}, {qp.proposal.bottle_slug for qp in pending})
@@ -97,18 +107,14 @@ class TestDiscoverPending(_FakeHomeMixin, unittest.TestCase):
now=datetime(2026, 5, 25, 14, 0, 0, tzinfo=timezone.utc),
)
for p in (late, early):
qdir = supervise.queue_dir_for_slug(p.bottle_slug)
qdir.mkdir(parents=True, exist_ok=True)
supervise.write_proposal(qdir, p)
supervise.write_proposal(p)
pending = supervise_cli.discover_pending()
self.assertEqual([early.id, late.id], [qp.proposal.id for qp in pending])
def test_excludes_already_responded(self):
p = _proposal()
qdir = supervise.queue_dir_for_slug("dev")
qdir.mkdir(parents=True)
supervise.write_proposal(qdir, p)
supervise.write_response(qdir, supervise.Response(
supervise.write_proposal(p)
supervise.write_response("dev", supervise.Response(
proposal_id=p.id, status=STATUS_APPROVED, notes="",
))
self.assertEqual([], supervise_cli.discover_pending())
@@ -123,10 +129,8 @@ class TestApproveReject(_FakeHomeMixin, unittest.TestCase):
def _enqueue(self, tool: str = TOOL_EGRESS_ALLOW):
p = _proposal(tool=tool)
qdir = supervise.queue_dir_for_slug("dev")
qdir.mkdir(parents=True, exist_ok=True)
supervise.write_proposal(qdir, p)
return supervise_cli.QueuedProposal(proposal=p, queue_dir=qdir)
supervise.write_proposal(p)
return supervise_cli.QueuedProposal(proposal=p)
def test_approve_writes_response(self):
qp = self._enqueue()
@@ -135,7 +139,7 @@ class TestApproveReject(_FakeHomeMixin, unittest.TestCase):
return_value=("routes: []\n", "routes:\n - host: example.com\n"),
):
supervise_cli.approve(qp)
resp = read_response(qp.queue_dir, qp.proposal.id)
resp = read_response(qp.proposal.bottle_slug, qp.proposal.id)
self.assertEqual(STATUS_APPROVED, resp.status)
self.assertIsNone(resp.final_file)
@@ -150,7 +154,7 @@ class TestApproveReject(_FakeHomeMixin, unittest.TestCase):
final_file="routes:\n - host: edited.example.com\n",
notes="tweaked",
)
resp = read_response(qp.queue_dir, qp.proposal.id)
resp = read_response(qp.proposal.bottle_slug, qp.proposal.id)
self.assertEqual(STATUS_MODIFIED, resp.status)
self.assertEqual("routes:\n - host: edited.example.com\n", resp.final_file)
self.assertEqual("tweaked", resp.notes)
@@ -158,7 +162,7 @@ class TestApproveReject(_FakeHomeMixin, unittest.TestCase):
def test_reject_writes_rejection(self):
qp = self._enqueue()
supervise_cli.reject(qp, reason="nope")
resp = read_response(qp.queue_dir, qp.proposal.id)
resp = read_response(qp.proposal.bottle_slug, qp.proposal.id)
self.assertEqual(STATUS_REJECTED, resp.status)
self.assertEqual("nope", resp.notes)
@@ -181,36 +185,33 @@ class TestApproveReject(_FakeHomeMixin, unittest.TestCase):
def test_approve_gitleaks_allow_leaves_response_for_gate(self):
qp = self._enqueue(tool=TOOL_GITLEAKS_ALLOW)
supervise_cli.approve(qp, notes="dummy fixture")
# Gate polls the queue dir for the response; TUI must not archive it.
resp = read_response(qp.queue_dir, qp.proposal.id)
# Gate polls the DB for the response; TUI must not archive it.
resp = read_response(qp.proposal.bottle_slug, qp.proposal.id)
self.assertEqual(STATUS_APPROVED, resp.status)
self.assertEqual("dummy fixture", resp.notes)
self.assertFalse((qp.queue_dir / "processed").exists())
def test_tui_gitleaks_allow_requires_reason(self):
qp = self._enqueue(tool=TOOL_GITLEAKS_ALLOW)
with patch.object(supervise_cli, "_prompt", return_value=""):
status = supervise_cli._approve_from_tui(None, qp) # type: ignore[arg-type]
self.assertEqual("approve aborted (empty reason)", status)
self.assertFalse((qp.queue_dir / "processed").exists())
def test_tui_gitleaks_allow_writes_reason(self):
qp = self._enqueue(tool=TOOL_GITLEAKS_ALLOW)
with patch.object(supervise_cli, "_prompt", return_value="test fixture"):
status = supervise_cli._approve_from_tui(None, qp) # type: ignore[arg-type]
self.assertIn("approved gitleaks-allow", status)
resp = read_response(qp.queue_dir, qp.proposal.id)
resp = read_response(qp.proposal.bottle_slug, qp.proposal.id)
self.assertEqual("test fixture", resp.notes)
def test_approve_token_allow_leaves_response_for_egress(self):
qp = self._enqueue(tool=TOOL_EGRESS_TOKEN_ALLOW)
supervise_cli.approve(qp, notes="false positive")
# The egress addon polls the queue dir for the response; the TUI must
# The egress addon polls the DB for the response; the TUI must
# not archive it (the addon archives after reading).
resp = read_response(qp.queue_dir, qp.proposal.id)
resp = read_response(qp.proposal.bottle_slug, qp.proposal.id)
self.assertEqual(STATUS_APPROVED, resp.status)
self.assertEqual("false positive", resp.notes)
self.assertFalse((qp.queue_dir / "processed").exists())
def test_token_allow_writes_no_audit_log(self):
qp = self._enqueue(tool=TOOL_EGRESS_TOKEN_ALLOW)
@@ -222,14 +223,13 @@ class TestApproveReject(_FakeHomeMixin, unittest.TestCase):
with patch.object(supervise_cli, "_prompt", return_value=""):
status = supervise_cli._approve_from_tui(None, qp) # type: ignore[arg-type]
self.assertEqual("approve aborted (empty reason)", status)
self.assertFalse((qp.queue_dir / "processed").exists())
def test_tui_token_allow_writes_reason(self):
qp = self._enqueue(tool=TOOL_EGRESS_TOKEN_ALLOW)
with patch.object(supervise_cli, "_prompt", return_value="legit"):
status = supervise_cli._approve_from_tui(None, qp) # type: ignore[arg-type]
self.assertIn("approved egress-token-allow", status)
resp = read_response(qp.queue_dir, qp.proposal.id)
resp = read_response(qp.proposal.bottle_slug, qp.proposal.id)
self.assertEqual("legit", resp.notes)
def test_suffix_for_token_allow_is_txt(self):
+126 -54
View File
@@ -4,7 +4,6 @@ fallback paths."""
from __future__ import annotations
import os
import tempfile
import time
import unittest
@@ -12,14 +11,19 @@ from pathlib import Path
from unittest.mock import patch
from bot_bottle import supervise
from bot_bottle.audit_store import AuditStore
from bot_bottle.queue_store import QueueStore
from bot_bottle.supervise import (
AuditEntry,
Proposal,
STATUS_APPROVED,
TOOL_EGRESS_ALLOW,
list_pending_proposals,
read_audit_entries,
read_proposal,
read_response,
wait_for_response,
write_audit_entry,
)
@@ -37,58 +41,59 @@ class TestPathHelpers(unittest.TestCase):
def test_bot_bottle_root(self) -> None:
self.assertTrue(str(supervise.bot_bottle_root()).endswith(".bot-bottle"))
def test_queue_dir_for_slug(self) -> None:
self.assertIn("slug", str(supervise.queue_dir_for_slug("slug")))
def test_id_from_non_proposal_filename(self) -> None:
self.assertIsNone(supervise._id_from_proposal_filename(Path("x.response.json")))
class TestReadMalformed(unittest.TestCase):
def test_read_proposal_non_dict(self) -> None:
def test_read_proposal_missing_row(self) -> None:
with tempfile.TemporaryDirectory() as d:
(Path(d) / "p.proposal.json").write_text("[]")
with self.assertRaises(ValueError):
read_proposal(Path(d), "p")
with patch.dict("os.environ", {"HOME": d}):
QueueStore("slug").migrate()
with self.assertRaises(FileNotFoundError):
read_proposal("slug", "p")
def test_read_response_non_dict(self) -> None:
def test_read_response_missing_row(self) -> None:
with tempfile.TemporaryDirectory() as d:
(Path(d) / "p.response.json").write_text("[]")
with self.assertRaises(ValueError):
read_response(Path(d), "p")
with patch.dict("os.environ", {"HOME": d}):
QueueStore("slug").migrate()
with self.assertRaises(FileNotFoundError):
read_response("slug", "p")
def test_list_pending_skips_malformed(self) -> None:
def test_list_pending_reads_db_only(self) -> None:
with tempfile.TemporaryDirectory() as d:
qd = Path(d)
(qd / "bad.proposal.json").write_text("{ not json")
(qd / "arr.proposal.json").write_text("[]")
(qd / "incomplete.proposal.json").write_text("{}") # from_dict raises
supervise.write_proposal(qd, _proposal()) # one valid
pending = list_pending_proposals(qd)
with patch.dict("os.environ", {"HOME": d}):
QueueStore("slug").migrate()
supervise.write_proposal(_proposal())
pending = list_pending_proposals("slug")
self.assertEqual(1, len(pending))
self.assertEqual("slug", pending[0].bottle_slug)
def test_list_pending_skips_when_response_present(self) -> None:
with tempfile.TemporaryDirectory() as d:
qd = Path(d)
with patch.dict("os.environ", {"HOME": d}):
QueueStore("slug").migrate()
p = _proposal()
supervise.write_proposal(qd, p)
(qd / f"{p.id}.response.json").write_text("{}") # response exists -> skipped
self.assertEqual([], list_pending_proposals(qd))
supervise.write_proposal(p)
supervise.write_response("slug", supervise.Response(
proposal_id=p.id,
status=STATUS_APPROVED,
notes="",
))
self.assertEqual([], list_pending_proposals("slug"))
class TestWaitForResponse(unittest.TestCase):
def test_malformed_response_then_timeout(self) -> None:
def test_missing_response_times_out(self) -> None:
with tempfile.TemporaryDirectory() as d:
(Path(d) / "p.response.json").write_text("{ not json")
with patch.dict("os.environ", {"HOME": d}):
QueueStore("slug").migrate()
with self.assertRaises(TimeoutError):
wait_for_response(Path(d), "p", deadline=time.monotonic())
wait_for_response("slug", "p", deadline=time.monotonic())
def test_incomplete_response_then_timeout(self) -> None:
def test_empty_db_response_does_not_count(self) -> None:
with tempfile.TemporaryDirectory() as d:
(Path(d) / "p.response.json").write_text("{}") # dict but from_dict raises
with patch.dict("os.environ", {"HOME": d}):
QueueStore("slug").migrate()
with self.assertRaises(TimeoutError):
wait_for_response(Path(d), "p", deadline=time.monotonic())
wait_for_response("slug", "p", deadline=time.monotonic())
class TestReadAuditEntries(unittest.TestCase):
@@ -97,35 +102,102 @@ class TestReadAuditEntries(unittest.TestCase):
patch.dict("os.environ", {"HOME": home}):
self.assertEqual([], read_audit_entries("egress", "nope"))
def test_skips_malformed_lines(self) -> None:
def test_reads_entries_from_db(self) -> None:
with tempfile.TemporaryDirectory() as home, \
patch.dict("os.environ", {"HOME": home}):
path = supervise.audit_log_path("egress", "slug")
path.parent.mkdir(parents=True, exist_ok=True)
valid = (
'{"timestamp": "t", "bottle_slug": "slug", "component": "egress",'
' "operator_action": "approve", "operator_notes": "",'
' "justification": "", "diff": ""}'
)
path.write_text(
"\n" # blank line skipped
"{ not json\n" # JSONDecodeError skipped
"[]\n" # not a dict skipped
"{}\n" # missing fields -> ValueError skipped
+ valid + "\n"
)
AuditStore().migrate()
write_audit_entry(AuditEntry(
timestamp="t",
bottle_slug="slug",
component="egress",
operator_action="approve",
operator_notes="",
justification="",
diff="",
))
write_audit_entry(AuditEntry(
timestamp="t",
bottle_slug="other",
component="egress",
operator_action="reject",
operator_notes="",
justification="",
diff="",
))
entries = read_audit_entries("egress", "slug")
self.assertEqual(1, len(entries))
self.assertEqual("approve", entries[0].operator_action)
def test_legacy_audit_log_file_does_not_count(self) -> None:
with tempfile.TemporaryDirectory() as home, \
patch.dict("os.environ", {"HOME": home}):
path = supervise.audit_log_path("egress", "slug")
path.parent.mkdir(parents=True, exist_ok=True)
path.write_text(
'{"timestamp": "t", "bottle_slug": "slug", "component": "egress",'
' "operator_action": "approve", "operator_notes": "",'
' "justification": "", "diff": ""}\n'
)
entries = read_audit_entries("egress", "slug")
self.assertEqual([], entries)
class TestFlockFallback(unittest.TestCase):
def test_flock_on_closed_fd_is_swallowed(self) -> None:
# flock on a closed fd raises OSError(EBADF), which the helpers swallow.
fd = os.open(os.devnull, os.O_RDONLY)
os.close(fd)
supervise._try_flock(fd)
supervise._try_funlock(fd)
class TestStoreGuardBranches(unittest.TestCase):
"""Direct QueueStore / AuditStore construction and early-return guard branches."""
def test_queue_store_explicit_db_path(self):
with tempfile.TemporaryDirectory() as d:
db = Path(d) / "q.db"
store = QueueStore("key", db_path=db)
store.migrate()
self.assertTrue(db.is_file())
self.assertEqual(db, store.db_path)
def test_queue_store_missing_db_list_pending_returns_empty(self):
with tempfile.TemporaryDirectory() as d:
db = Path(d) / "q.db"
store = QueueStore("key", db_path=db)
store.migrate()
db.unlink()
self.assertEqual([], store.list_pending_proposals())
def test_queue_store_missing_db_list_all_returns_empty(self):
with tempfile.TemporaryDirectory() as d:
db = Path(d) / "q.db"
store = QueueStore("key", db_path=db)
store.migrate()
db.unlink()
self.assertEqual([], store.list_all_pending_proposals())
def test_queue_store_missing_db_archive_is_noop(self):
with tempfile.TemporaryDirectory() as d:
db = Path(d) / "q.db"
store = QueueStore("key", db_path=db)
store.migrate()
db.unlink()
store.archive_proposal("anything") # must not raise
def test_queue_store_chmod_oserror_is_swallowed(self):
with tempfile.TemporaryDirectory() as d:
db = Path(d) / "q.db"
store = QueueStore("key", db_path=db)
with patch("pathlib.Path.chmod", side_effect=OSError("ro")):
store.migrate() # must not raise
def test_audit_store_missing_db_read_returns_empty(self):
with tempfile.TemporaryDirectory() as d:
db = Path(d) / "a.db"
store = AuditStore(db_path=db)
store.migrate()
db.unlink()
self.assertEqual([], store.read_audit_entries("egress", "slug"))
def test_audit_store_chmod_oserror_is_swallowed(self):
with tempfile.TemporaryDirectory() as d:
db = Path(d) / "a.db"
store = AuditStore(db_path=db)
with patch("pathlib.Path.chmod", side_effect=OSError("ro")):
store.migrate() # must not raise
if __name__ == "__main__":
+38 -18
View File
@@ -17,8 +17,12 @@ from unittest.mock import patch
# bare name `supervise`.
sys.path.insert(0, str(Path(__file__).resolve().parent.parent.parent / "bot_bottle"))
import supervise as _sv # noqa: E402 # type: ignore
import queue_store as _qs # noqa: E402 # type: ignore
import audit_store as _as # noqa: E402 # type: ignore
import supervise_types as svt_flat # noqa: E402 # type: ignore
from bot_bottle import supervise_server # noqa: E402
from bot_bottle import supervise_types as svt_pkg # noqa: E402
from bot_bottle.supervise_server import (
ERR_INTERNAL,
ERR_INVALID_PARAMS,
@@ -112,7 +116,7 @@ class TestRpcErrorTaxonomy(unittest.TestCase):
validate_proposed_file(_sv.TOOL_EGRESS_ALLOW, "routes: nope\n")
def test_unknown_tool_in_tools_call_is_client_error(self):
config = ServerConfig(bottle_slug="dev", queue_dir=Path("/unused"))
config = ServerConfig(bottle_slug="dev")
with self.assertRaises(_RpcClientError) as cm:
handle_tools_call({"name": "no-such-tool", "arguments": {}}, config)
self.assertEqual(ERR_INVALID_PARAMS, cm.exception.code)
@@ -122,9 +126,9 @@ class TestRpcInternalErrorOnIoFailure(unittest.TestCase):
def test_write_proposal_os_error_raises_internal(self):
config = ServerConfig(
bottle_slug="dev",
queue_dir=Path("/dev/null/cannot-exist"),
)
with self.assertRaises(_RpcInternalError) as cm:
with patch.object(_sv, "write_proposal", side_effect=OSError("disk full")), \
self.assertRaises(_RpcInternalError) as cm:
handle_tools_call(
{
"name": _sv.TOOL_EGRESS_ALLOW,
@@ -265,21 +269,43 @@ class TestHandleToolsList(unittest.TestCase):
class TestHandleToolsCall(unittest.TestCase):
def setUp(self):
self._tmp = tempfile.TemporaryDirectory(prefix="supervise-server-test.")
self.queue_dir = Path(self._tmp.name)
self.config = ServerConfig(bottle_slug="dev", queue_dir=self.queue_dir)
self._home_patch = self._patch_home(Path(self._tmp.name))
self.config = ServerConfig(bottle_slug="dev")
_qs.QueueStore("dev").migrate()
_as.AuditStore().migrate()
def tearDown(self):
self._home_patch()
self._tmp.cleanup()
def _patch_home(self, fake_home: Path):
original_sv = _sv.bot_bottle_root
original_flat = svt_flat.bot_bottle_root
original_pkg = svt_pkg.bot_bottle_root
def fake_root() -> Path:
return fake_home / ".bot-bottle"
_sv.bot_bottle_root = fake_root # type: ignore[assignment]
svt_flat.bot_bottle_root = fake_root # type: ignore[assignment]
svt_pkg.bot_bottle_root = fake_root # type: ignore[assignment]
def restore() -> None:
_sv.bot_bottle_root = original_sv # type: ignore[assignment]
svt_flat.bot_bottle_root = original_flat # type: ignore[assignment]
svt_pkg.bot_bottle_root = original_pkg # type: ignore[assignment]
return restore
def _respond_when_proposal_appears(self, status: str, notes: str = "") -> threading.Thread:
"""Background thread: poll the queue for a fresh proposal, write a
matching response. Returns the thread so the test can join it."""
def runner():
for _ in range(200):
pending = _sv.list_pending_proposals(self.queue_dir)
pending = _sv.list_pending_proposals("dev")
if pending:
p = pending[0]
_sv.write_response(self.queue_dir, _sv.Response(
_sv.write_response("dev", _sv.Response(
proposal_id=p.id, status=status, notes=notes,
))
return
@@ -412,15 +438,11 @@ class TestHandleToolsCall(unittest.TestCase):
finally:
responder.join()
# No pending proposals left after archive.
self.assertEqual([], _sv.list_pending_proposals(self.queue_dir))
# Both files moved to processed/.
processed = list((self.queue_dir / "processed").glob("*.json"))
self.assertEqual(2, len(processed))
self.assertEqual([], _sv.list_pending_proposals("dev"))
def test_pending_response_times_out_without_archive(self):
config = ServerConfig(
bottle_slug="dev",
queue_dir=self.queue_dir,
response_timeout_seconds=0.05,
)
result = handle_tools_call(
@@ -438,8 +460,7 @@ class TestHandleToolsCall(unittest.TestCase):
text = result["content"][0]["text"] # type: ignore[index]
self.assertIn("status: pending", text)
self.assertIn("proposal remains queued", text)
self.assertEqual(1, len(_sv.list_pending_proposals(self.queue_dir)))
self.assertFalse((self.queue_dir / "processed").exists())
self.assertEqual(1, len(_sv.list_pending_proposals("dev")))
class TestHandleListEgressRoutes(unittest.TestCase):
@@ -461,7 +482,7 @@ class TestHandleListEgressRoutes(unittest.TestCase):
with patch.object(supervise_server.urllib.request, "build_opener", return_value=_Opener()):
result = handle_list_egress_routes(
{},
ServerConfig(bottle_slug="dev", queue_dir=Path("/unused")),
ServerConfig(bottle_slug="dev"),
)
self.assertFalse(result["isError"]) # type: ignore[index]
@@ -476,7 +497,7 @@ class TestHandleListEgressRoutes(unittest.TestCase):
with patch.object(supervise_server.urllib.request, "build_opener", return_value=_Opener()):
result = handle_list_egress_routes(
{},
ServerConfig(bottle_slug="dev", queue_dir=Path("/unused")),
ServerConfig(bottle_slug="dev"),
)
self.assertTrue(result["isError"]) # type: ignore[index]
@@ -544,7 +565,6 @@ class TestHttpEndToEnd(unittest.TestCase):
def setUp(self):
self._tmp = tempfile.TemporaryDirectory(prefix="supervise-http-test.")
self.queue_dir = Path(self._tmp.name)
# Pick a random port by binding to :0 first.
import socket
s = socket.socket()
@@ -552,7 +572,7 @@ class TestHttpEndToEnd(unittest.TestCase):
self.port = s.getsockname()[1]
s.close()
self.server = MCPServer(("127.0.0.1", self.port), MCPHandler)
self.server.config = ServerConfig(bottle_slug="dev", queue_dir=self.queue_dir)
self.server.config = ServerConfig(bottle_slug="dev")
self.thread = threading.Thread(
target=self.server.serve_forever, daemon=True,
)