fix(supervise): reach the queue over RPC, get bot-bottle.db off the data plane

PRD 0070's rule — only the orchestrator opens bot-bottle.db; the data
plane reaches state through the control-plane RPC — was not in force.
Three data-plane daemons held a direct read-write handle on the shared
SQLite file: the supervise MCP server, the egress DLP addon (the most
attack-exposed process, TLS-bumping hostile traffic), and the git-gate
pre-receive hook. An RCE in any of them could read every bottle's
plaintext identity_token and forge attribution fleet-wide (issue #469).

Add the agent half of the supervise flow to the control plane:

  POST /supervise/propose  -> queue a proposal, 201 {proposal_id}
  POST /supervise/poll     -> non-blocking decision poll, 200 {status,...}

Both attribute the caller by (source_ip, identity_token) exactly like
/resolve — never a caller-supplied slug — so a bottle can only ever queue
or read its own proposals even if the data plane is compromised. A decided
poll archives server-side, preserving the archive-after-read contract.

Data plane: the supervise server, egress addon, and git-gate hook now
queue/poll through PolicyResolver.propose_supervise / poll_supervise
instead of opening the DB. supervise_server keeps its ~30s grace window
by polling the RPC; egress keeps its safelist keyed by resolved bottle;
the git-gate hook gets (source_ip, identity_token) from the CGI env.

Packaging: drop the DB bind-mount and SUPERVISE_DB_PATH from the
data-plane containers/VMs (docker gateway + infra, macOS infra,
firecracker infra). The orchestrator remains the sole opener of the one
file via BOT_BOTTLE_ROOT / host_db_path().

Update PRD 0070: the rule is now in force; remove the transitional caveat.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
This commit is contained in:
2026-07-24 02:07:21 +00:00
parent f24ae45d13
commit 2c496dc3d0
23 changed files with 968 additions and 518 deletions
+44 -37
View File
@@ -272,21 +272,27 @@ supervise_gitleaks_allow() {
proposal_id=$(
PYTHONPATH="/app${PYTHONPATH:+:$PYTHONPATH}" GITLEAKS_ALLOW_REF="$ref" python3 - "$report_file" <<'PY'
import datetime
import hashlib
import json
import os
import sys
from pathlib import Path
# Queue over the control-plane RPC — the git-gate data plane no longer opens
# bot-bottle.db (PRD 0070 / issue #469). Attribution is by (source_ip,
# identity_token), resolved server-side, so the proposal lands under the
# calling bottle exactly as a direct write once did.
try:
import supervise as _sv
from bot_bottle.policy_resolver import PolicyResolver, PolicyResolveError
from bot_bottle.supervise_types import TOOL_GITLEAKS_ALLOW
except ImportError:
from bot_bottle import supervise as _sv
from policy_resolver import PolicyResolver, PolicyResolveError
from supervise_types import TOOL_GITLEAKS_ALLOW
report_path = Path(sys.argv[1])
slug = os.environ.get("SUPERVISE_BOTTLE_SLUG", "")
if not slug:
source_ip = os.environ.get("SUPERVISE_SOURCE_IP", "")
token = os.environ.get("SUPERVISE_IDENTITY_TOKEN", "")
orch_url = os.environ.get("BOT_BOTTLE_ORCHESTRATOR_URL", "")
if not source_ip or not orch_url:
sys.exit(2)
try:
@@ -323,19 +329,21 @@ for i, finding in enumerate(raw, 1):
])
payload = "\n".join(lines).rstrip() + "\n"
proposal = _sv.Proposal.new(
bottle_slug=slug,
tool=_sv.TOOL_GITLEAKS_ALLOW,
proposed_file=payload,
justification=(
"git-gate found gitleaks findings hidden by # gitleaks:allow; "
"approve only for dummy test fixtures or confirmed false positives"
),
current_file_hash=hashlib.sha256(payload.encode("utf-8")).hexdigest(),
now=datetime.datetime.now(datetime.timezone.utc),
)
_sv.write_proposal(proposal)
print(proposal.id)
try:
proposal_id = PolicyResolver(orch_url).propose_supervise(
source_ip, token,
tool=TOOL_GITLEAKS_ALLOW,
proposed_file=payload,
justification=(
"git-gate found gitleaks findings hidden by # gitleaks:allow; "
"approve only for dummy test fixtures or confirmed false positives"
),
)
except PolicyResolveError:
sys.exit(4)
if not proposal_id:
sys.exit(2)
print(proposal_id)
PY
)
rc=$?
@@ -348,7 +356,6 @@ PY
return 1
fi
slug=${SUPERVISE_BOTTLE_SLUG:-}
timeout=${SUPERVISE_GITLEAKS_ALLOW_TIMEOUT_SECONDS:-300}
case "$timeout" in
''|*[!0-9]*)
@@ -360,20 +367,30 @@ PY
echo "git-gate: approve with './cli.py supervise' to continue this push" >&2
waited=0
while [ "$waited" -lt "$timeout" ]; do
status=$(PYTHONPATH="/app${PYTHONPATH:+:$PYTHONPATH}" python3 - "$slug" "$proposal_id" <<'PY'
status=$(PYTHONPATH="/app${PYTHONPATH:+:$PYTHONPATH}" python3 - "$proposal_id" <<'PY'
import os
import sys
# Non-blocking poll over the control plane. A decided proposal is archived
# server-side on read, so no separate archive step is needed here.
try:
import supervise as _sv
from bot_bottle.policy_resolver import PolicyResolver, PolicyResolveError
except ImportError:
from bot_bottle import supervise as _sv
from policy_resolver import PolicyResolver, PolicyResolveError
slug = sys.argv[1]
source_ip = os.environ.get("SUPERVISE_SOURCE_IP", "")
token = os.environ.get("SUPERVISE_IDENTITY_TOKEN", "")
orch_url = os.environ.get("BOT_BOTTLE_ORCHESTRATOR_URL", "")
try:
response = _sv.read_response(slug, sys.argv[2])
except FileNotFoundError:
result = PolicyResolver(orch_url).poll_supervise(source_ip, token, sys.argv[1])
except PolicyResolveError:
sys.exit(2)
print(response.status)
if result is None:
sys.exit(2)
status = result.get("status", "")
if status in ("pending", "unknown"):
sys.exit(2)
print(status)
PY
)
rc=$?
@@ -385,16 +402,6 @@ PY
if [ -n "$status" ]; then
case "$status" in
approved|modified)
PYTHONPATH="/app${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
;;