docs(prd): 0070 expand the consolidated-sidecar arc in the roadmap (#352)

Split the sidecar work into explicit slices: lifecycle (done), build the
real bundle image (done), source-IP-keyed multi-tenant config +
registration/reload (the functional core, next), and routing agent bottles
through it. Renumber firecracker/macos to 8/9. Slices 1-5 implemented
(#352/#356/#357/#358/#360).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WBMWTEtQdJ4W5UrWuLHCck

Dockerfile.sidecars

@@ -66,6 +66,7 @@ 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/paths.py             /app/paths.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

bot_bottle/audit_store.py

@@ -6,11 +6,13 @@ import sqlite3
 from pathlib import Path
 
 try:
-    from .supervise_types import AuditEntry, host_db_path
+    from .supervise_types import AuditEntry
+    from .paths import 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 supervise_types import AuditEntry  # type: ignore[import-not-found]  # pylint: disable=import-error,no-name-in-module
+    from paths import 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
 

bot_bottle/backend/docker/cleanup.py

@@ -26,7 +26,7 @@ from __future__ import annotations
 import shutil
 import subprocess
 
-from ... import supervise as _supervise
+from ...paths import bot_bottle_root
 from ...log import info, warn
 from . import util as docker_mod
 from .bottle_cleanup_plan import DockerBottleCleanupPlan
@@ -94,7 +94,7 @@ def _list_orphan_state_dirs(
     ANY backend — used so this docker-side check doesn't reap a
     running non-docker bottle's state dir (the layout is shared
     across backends)."""
-    state_root = _supervise.bot_bottle_root() / "state"
+    state_root = bot_bottle_root() / "state"
     if not state_root.is_dir():
         return []
     orphans: list[str] = []

bot_bottle/bottle_state.py

@@ -36,7 +36,7 @@ from dataclasses import dataclass
 from pathlib import Path
 from typing import cast
 
-from . import supervise as _supervise
+from .paths import bot_bottle_root
 
 
 # Directory layout: ~/.bot-bottle/state/<identity>/...
@@ -163,7 +163,7 @@ def read_metadata(identity: str) -> BottleMetadata | None:
 def bottle_state_dir(identity: str) -> Path:
     """Per-bottle state directory on the host. Created lazily by the
     write helpers; readers tolerate its absence."""
-    return _supervise.bot_bottle_root() / _STATE_SUBDIR / identity
+    return bot_bottle_root() / _STATE_SUBDIR / identity
 
 
 def per_bottle_dockerfile_path(identity: str) -> Path:

bot_bottle/cli/supervise.py

@@ -19,7 +19,7 @@ from dataclasses import dataclass
 from datetime import datetime, timezone
 from pathlib import Path
 
-from .. import supervise as _supervise
+from ..paths import bot_bottle_root
 from ..bottle_state import read_metadata
 from ..backend.docker.egress_apply import (
     EgressApplyError,
@@ -276,7 +276,7 @@ def _write_crash_log(exc: BaseException) -> Path:
     )
     entry = f"=== supervise crash {stamp} ===\n{body}\n"
     try:
-        log_dir = _supervise.bot_bottle_root() / "logs"
+        log_dir = bot_bottle_root() / "logs"
         log_dir.mkdir(parents=True, exist_ok=True)
         path = log_dir / "supervise-crash.log"
         with path.open("a", encoding="utf-8") as fh:

bot_bottle/orchestrator/__init__.py

@@ -0,0 +1,30 @@
+"""Per-host orchestrator service (PRD 0070).
+
+A single persistent per-host service that will run the sidecar functions
+(egress / git-gate / supervise), coordinate with the console, and broker
+agent launches. This package is being built bottom-up, starting with the
+backend-neutral "consolidation core" that needs no VM packaging:
+
+  * `registry`      — the SQLite runtime-state store + fail-closed
+                      attribution (source IP + per-bottle identity token).
+  * `control_plane` — the HTTP control-plane RPC (register / deregister /
+                      list / attribute / health), with live reload.
+
+Launch/teardown, the launch broker, and the actual egress/git/supervise
+data plane land in later slices once this core is proven (see the PRD's
+sequencing: plain-process dev-harness -> docker orchestrator -> firecracker).
+"""
+
+from __future__ import annotations
+
+from .registry import BottleRecord, RegistryStore, new_identity_token
+from .control_plane import ControlPlaneServer, dispatch, make_server
+
+__all__ = [
+    "BottleRecord",
+    "RegistryStore",
+    "new_identity_token",
+    "ControlPlaneServer",
+    "dispatch",
+    "make_server",
+]

bot_bottle/orchestrator/__main__.py

@@ -0,0 +1,52 @@
+"""Run the orchestrator control plane as a plain process (PRD 0070 dev-harness).
+
+    python -m bot_bottle.orchestrator [--host H] [--port P] [--db PATH]
+
+The PRD sequences the orchestrator as a plain-process dev-harness first, so
+the consolidation core (registry + attribution + HTTP control plane + live
+reload) can be exercised with fast iteration, decoupled from any VM /
+container packaging. Wrapping this exact service in a backend-native unit
+(docker container, then Firecracker VM) comes later.
+"""
+
+from __future__ import annotations
+
+import argparse
+from pathlib import Path
+
+from .. import log
+from .control_plane import make_server
+from .registry import RegistryStore, default_db_path
+
+
+def main(argv: list[str] | None = None) -> int:
+    """Parse args, migrate the registry, and serve the control plane."""
+    parser = argparse.ArgumentParser(prog="bot_bottle.orchestrator")
+    parser.add_argument("--host", default="127.0.0.1", help="bind address")
+    parser.add_argument("--port", type=int, default=8080, help="bind port (0 = ephemeral)")
+    parser.add_argument(
+        "--db", type=Path, default=None,
+        help=f"registry DB path (default: {default_db_path()})",
+    )
+    args = parser.parse_args(argv)
+
+    registry = RegistryStore(args.db)
+    registry.migrate()
+
+    server = make_server(registry, host=args.host, port=args.port)
+    bound_host, bound_port = server.server_address[0], server.server_address[1]
+    log.info(
+        "orchestrator control plane listening",
+        context={"host": bound_host, "port": bound_port, "db": str(registry.db_path)},
+    )
+    try:
+        server.serve_forever()
+    except KeyboardInterrupt:
+        log.info("orchestrator shutting down")
+    finally:
+        server.server_close()
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

bot_bottle/orchestrator/control_plane.py

@@ -0,0 +1,154 @@
+"""Orchestrator HTTP control plane (PRD 0070).
+
+The backend-agnostic control-plane RPC (CLI / console -> orchestrator) over
+**HTTP** — the universal transport chosen in 0070 (works on every host; no
+vsock / unix-socket portability caveats). Endpoints mutate the live
+registry, so register/deregister are the *live-reload* path — no restart:
+
+    GET    /health              -> 200 {"status": "ok"}
+    GET    /bottles             -> 200 {"bottles": [ <redacted record>, ... ]}
+    POST   /bottles             -> 201 {"bottle_id", "identity_token"}
+                                   body: {"source_ip", ["bottle_id"], ["metadata"]}
+    DELETE /bottles/<bottle_id> -> 200 {"deregistered": true} | 404
+    POST   /attribute           -> 200 {"bottle_id"} | 403
+                                   body: {"source_ip", "identity_token"}
+
+Routing/handling is the pure function `dispatch()` so it is unit-testable
+without a socket; `Handler` / `ControlPlaneServer` / `make_server` are a
+thin stdlib adapter around it.
+
+Note the listing redacts identity tokens — they are never returned except
+once, to the caller that registers the bottle.
+"""
+
+from __future__ import annotations
+
+import http.server
+import json
+import os
+import socketserver
+import typing
+from urllib.parse import urlsplit
+
+from .registry import RegistryStore
+
+# JSON body payload type (parsed request / rendered response).
+Json = dict[str, object]
+
+
+def _parse_json_object(body: bytes) -> Json:
+    """Parse a JSON object body. Raises ValueError for non-objects / bad JSON."""
+    if not body:
+        return {}
+    obj = json.loads(body)  # raises json.JSONDecodeError (a ValueError)
+    if not isinstance(obj, dict):
+        raise ValueError("request body must be a JSON object")
+    return obj
+
+
+def dispatch(  # pylint: disable=too-many-return-statements
+    registry: RegistryStore, method: str, path: str, body: bytes
+) -> tuple[int, Json]:
+    """Route one control-plane request to a (status, payload) pair. Pure —
+    no I/O beyond the registry — so it is fully testable without a socket."""
+    route = urlsplit(path).path.rstrip("/") or "/"
+
+    if method == "GET" and route == "/health":
+        return 200, {"status": "ok"}
+
+    if method == "GET" and route == "/bottles":
+        return 200, {"bottles": [r.redacted() for r in registry.all()]}
+
+    if method == "POST" and route == "/bottles":
+        try:
+            data = _parse_json_object(body)
+        except ValueError as e:
+            return 400, {"error": f"invalid JSON: {e}"}
+        source_ip = data.get("source_ip")
+        if not isinstance(source_ip, str) or not source_ip:
+            return 400, {"error": "source_ip (string) is required"}
+        bottle_id = data.get("bottle_id")
+        metadata = data.get("metadata")
+        rec = registry.register(
+            source_ip,
+            bottle_id=bottle_id if isinstance(bottle_id, str) else None,
+            metadata=metadata if isinstance(metadata, str) else "",
+        )
+        return 201, {"bottle_id": rec.bottle_id, "identity_token": rec.identity_token}
+
+    if method == "DELETE" and route.startswith("/bottles/"):
+        bottle_id = route[len("/bottles/"):]
+        if registry.deregister(bottle_id):
+            return 200, {"deregistered": True}
+        return 404, {"error": "no such bottle"}
+
+    if method == "POST" and route == "/attribute":
+        try:
+            data = _parse_json_object(body)
+        except ValueError as e:
+            return 400, {"error": f"invalid JSON: {e}"}
+        source_ip = data.get("source_ip")
+        token = data.get("identity_token")
+        if not isinstance(source_ip, str) or not isinstance(token, str):
+            return 400, {"error": "source_ip and identity_token (strings) required"}
+        rec = registry.attribute(source_ip, token)
+        if rec is None:
+            return 403, {"error": "unattributed"}
+        return 200, {"bottle_id": rec.bottle_id}
+
+    return 404, {"error": "not found"}
+
+
+class Handler(http.server.BaseHTTPRequestHandler):
+    """Thin stdlib adapter: read the body, call `dispatch`, write JSON."""
+
+    # Quiet by default (the orchestrator has its own logging); opt back into
+    # stdlib access logging with BOT_BOTTLE_ORCHESTRATOR_DEBUG.
+    def log_message(self, format: str, *args: typing.Any) -> None:  # noqa: A002
+        if os.environ.get("BOT_BOTTLE_ORCHESTRATOR_DEBUG"):
+            super().log_message(format, *args)
+
+    def _serve(self, method: str) -> None:
+        """Read the request body, dispatch it, and write the JSON reply."""
+        server = self.server
+        assert isinstance(server, ControlPlaneServer)
+        length = int(self.headers.get("Content-Length") or 0)
+        body = self.rfile.read(length) if length > 0 else b""
+        status, payload = dispatch(server.registry, method, self.path, body)
+        data = json.dumps(payload).encode()
+        self.send_response(status)
+        self.send_header("Content-Type", "application/json")
+        self.send_header("Content-Length", str(len(data)))
+        self.end_headers()
+        self.wfile.write(data)
+
+    def do_GET(self) -> None:
+        self._serve("GET")
+
+    def do_POST(self) -> None:
+        self._serve("POST")
+
+    def do_DELETE(self) -> None:
+        self._serve("DELETE")
+
+
+class ControlPlaneServer(socketserver.ThreadingMixIn, http.server.HTTPServer):
+    """Threading HTTP server that carries the registry for its handlers."""
+
+    daemon_threads = True
+    allow_reuse_address = True
+
+    def __init__(self, address: tuple[str, int], registry: RegistryStore) -> None:
+        self.registry = registry
+        super().__init__(address, Handler)
+
+
+def make_server(
+    registry: RegistryStore, host: str = "127.0.0.1", port: int = 0
+) -> ControlPlaneServer:
+    """Build (but do not start) a control-plane server. `port=0` binds an
+    ephemeral port — read `server.server_address` for the actual one."""
+    return ControlPlaneServer((host, port), registry)
+
+
+__all__ = ["dispatch", "Handler", "ControlPlaneServer", "make_server", "Json"]

bot_bottle/orchestrator/registry.py

@@ -0,0 +1,219 @@
+"""Orchestrator bottle registry — the per-host runtime-state store (PRD 0070).
+
+SQLite-backed registry mapping each live bottle to its source IP and
+per-bottle identity token, plus the **fail-closed attribution** the data
+plane relies on: a request is attributed to a bottle only when its source
+IP *and* its identity token both match a single active record.
+
+The registry co-tenants the shared host `bot-bottle.db` (the `DbStore`
+framework namespaces each store by `schema_key`), so all bot-bottle
+runtime state lives in one queryable file — one place to back up, inspect,
+and integrate a console against.
+
+This is the "runtime state" tier of PRD 0070 (leases / approvals /
+registry) — deliberately NOT config (which stays declarative under
+`~/.bot-bottle/`) and NOT the build-time constants (a flat file). It is the
+source of truth the orchestrator sweeps on restart to re-adopt running
+bottles, so it lives in a durable store rather than process memory.
+
+Attribution combines two independent signals, and needs both:
+
+  * source IP  — the network-layer invariant (on Firecracker the `/31` TAP
+    + nft make it unspoofable by construction; weaker on other backends).
+  * identity token — an application-layer per-bottle secret, defence in
+    depth that doesn't lean on network anti-spoof. A bottle can only ever
+    prove it is *itself* (it can't learn another bottle's token), so a
+    hostile agent gains nothing by presenting it.
+"""
+
+from __future__ import annotations
+
+import hmac
+import secrets
+import sqlite3
+import time
+from dataclasses import dataclass
+from pathlib import Path
+
+from ..db_store import DbStore
+from ..migrations import TableMigrations
+from ..paths import host_db_path
+
+# 256 bits of urandom, URL-safe — unguessable per-bottle identity token.
+IDENTITY_TOKEN_BYTES = 32
+
+
+def new_identity_token() -> str:
+    """A fresh per-bottle identity token (PRD 0070 attribution defence)."""
+    return secrets.token_urlsafe(IDENTITY_TOKEN_BYTES)
+
+
+def default_db_path() -> Path:
+    """The shared host state DB (`bot-bottle.db`) — all bot-bottle runtime
+    state in one file, co-tenanted via schema_key. Host-resident so it
+    survives orchestrator restarts (re-adoption sweeps it)."""
+    return host_db_path()
+
+
+@dataclass(frozen=True)
+class BottleRecord:
+    """One live bottle in the registry."""
+
+    bottle_id: str
+    source_ip: str
+    identity_token: str
+    state: str = "active"
+    created_at: float = 0.0
+    # Opaque JSON blob for forward-compat (policy refs, pool slot, ...) —
+    # the registry doesn't interpret it, so new fields need no migration.
+    metadata: str = ""
+
+    def redacted(self) -> dict[str, object]:
+        """Public view for listing — never exposes the identity token."""
+        return {
+            "bottle_id": self.bottle_id,
+            "source_ip": self.source_ip,
+            "state": self.state,
+            "created_at": self.created_at,
+            "metadata": self.metadata,
+        }
+
+
+_MIGRATIONS = TableMigrations(
+    "orchestrator_registry",
+    [
+        # v1 — the live bottle registry.
+        """
+        CREATE TABLE IF NOT EXISTS orchestrator_bottles (
+            bottle_id      TEXT PRIMARY KEY,
+            source_ip      TEXT NOT NULL,
+            identity_token TEXT NOT NULL,
+            state          TEXT NOT NULL DEFAULT 'active',
+            created_at     REAL NOT NULL,
+            metadata       TEXT NOT NULL DEFAULT ''
+        )
+        """,
+        # source_ip is the data-plane attribution key — index it, and make
+        # the "one active bottle per source IP" check cheap.
+        "CREATE INDEX IF NOT EXISTS idx_orchestrator_bottles_source_ip "
+        "ON orchestrator_bottles (source_ip)",
+    ],
+)
+
+
+def _row_to_record(row: sqlite3.Row) -> BottleRecord:
+    """Build a BottleRecord from a registry row."""
+    return BottleRecord(
+        bottle_id=row["bottle_id"],
+        source_ip=row["source_ip"],
+        identity_token=row["identity_token"],
+        state=row["state"],
+        created_at=row["created_at"],
+        metadata=row["metadata"],
+    )
+
+
+class RegistryStore(DbStore):
+    """SQLite-backed registry of live bottles + fail-closed attribution."""
+
+    def __init__(self, db_path: Path | None = None) -> None:
+        super().__init__(db_path or default_db_path(), _MIGRATIONS)
+
+    def _connect(self) -> sqlite3.Connection:
+        """Open a connection with a busy timeout for the shared DB."""
+        conn = super()._connect()
+        # The registry co-tenants the shared bot-bottle.db, so a busy_timeout
+        # rides out brief lock contention with the other stores. WAL for the
+        # shared DB is a deliberate future change (it affects supervise/audit
+        # and is finicky over guest shares) — not flipped here.
+        conn.execute("PRAGMA busy_timeout=5000")
+        return conn
+
+    def register(
+        self,
+        source_ip: str,
+        *,
+        bottle_id: str | None = None,
+        identity_token: str | None = None,
+        metadata: str = "",
+    ) -> BottleRecord:
+        """Register (or replace) a bottle. Mints a bottle_id / identity token
+        when not supplied. Live — this is the control-plane reload path."""
+        rec = BottleRecord(
+            bottle_id=bottle_id or secrets.token_hex(8),
+            source_ip=source_ip,
+            identity_token=identity_token or new_identity_token(),
+            state="active",
+            created_at=time.time(),
+            metadata=metadata,
+        )
+        with self._connect() as conn:
+            conn.execute(
+                "INSERT OR REPLACE INTO orchestrator_bottles "
+                "(bottle_id, source_ip, identity_token, state, created_at, metadata) "
+                "VALUES (?, ?, ?, ?, ?, ?)",
+                (
+                    rec.bottle_id,
+                    rec.source_ip,
+                    rec.identity_token,
+                    rec.state,
+                    rec.created_at,
+                    rec.metadata,
+                ),
+            )
+        self._chmod()
+        return rec
+
+    def deregister(self, bottle_id: str) -> bool:
+        """Remove a bottle. Returns True if a row was deleted."""
+        with self._connect() as conn:
+            cur = conn.execute(
+                "DELETE FROM orchestrator_bottles WHERE bottle_id = ?", (bottle_id,)
+            )
+            return cur.rowcount > 0
+
+    def get(self, bottle_id: str) -> BottleRecord | None:
+        """Return the bottle by id, or None if absent."""
+        with self._connect() as conn:
+            row = conn.execute(
+                "SELECT * FROM orchestrator_bottles WHERE bottle_id = ?", (bottle_id,)
+            ).fetchone()
+        return _row_to_record(row) if row else None
+
+    def all(self) -> list[BottleRecord]:
+        """Every registered bottle, oldest first."""
+        with self._connect() as conn:
+            rows = conn.execute(
+                "SELECT * FROM orchestrator_bottles ORDER BY created_at"
+            ).fetchall()
+        return [_row_to_record(r) for r in rows]
+
+    def attribute(self, source_ip: str, identity_token: str) -> BottleRecord | None:
+        """Fail-closed attribution. Returns the bottle only when exactly one
+        active record has this source IP AND its identity token matches
+        (constant-time). Either signal alone is insufficient: an unknown IP,
+        an ambiguous IP (more than one active bottle — a misconfiguration),
+        an empty token, or a token mismatch all deny."""
+        if not identity_token:
+            return None
+        with self._connect() as conn:
+            rows = conn.execute(
+                "SELECT * FROM orchestrator_bottles "
+                "WHERE source_ip = ? AND state = 'active'",
+                (source_ip,),
+            ).fetchall()
+        if len(rows) != 1:
+            return None
+        rec = _row_to_record(rows[0])
+        if not hmac.compare_digest(rec.identity_token, identity_token):
+            return None
+        return rec
+
+
+__all__ = [
+    "BottleRecord",
+    "RegistryStore",
+    "new_identity_token",
+    "default_db_path",
+    "IDENTITY_TOKEN_BYTES",
+]

bot_bottle/paths.py

@@ -0,0 +1,43 @@
+"""Foundational filesystem paths for bot-bottle.
+
+`bot_bottle_root()` is the app data root — state, queue, audit logs,
+git-gate keys, and the shared DB all live under it. It defaults to
+`~/.bot-bottle` and is overridable with the **`BOT_BOTTLE_ROOT`** env var.
+
+The env override is the single knob for redirecting the root: the test
+suite points it at a throwaway dir instead of monkey-patching the function
+(every module and every flat/package copy reads the same env var, so one
+override covers them all), and operators can relocate the root if needed.
+
+This module has no bot-bottle imports, so it is safe to import from any
+layer (and to COPY flat into the sidecar bundle).
+"""
+
+from __future__ import annotations
+
+import os
+from pathlib import Path
+
+# The single shared host state DB. All bot-bottle SQLite stores (supervise
+# queue, audit, the orchestrator registry) co-tenant this one file — the
+# TableMigrations schema_key namespaces each store's tables.
+HOST_DB_FILENAME = "bot-bottle.db"
+
+
+def bot_bottle_root() -> Path:
+    """The app data root — `$BOT_BOTTLE_ROOT` if set, else `~/.bot-bottle`."""
+    override = os.environ.get("BOT_BOTTLE_ROOT")
+    return Path(override) if override else Path.home() / ".bot-bottle"
+
+
+def host_db_path() -> Path:
+    """Path to the shared host state DB, `<root>/db/bot-bottle.db`.
+
+    Kept in its own `db/` subdirectory (not directly under the root) so a
+    backend that can only bind-mount *directories* can share this one file
+    with a sidecar without exposing the root's other contents (git-gate
+    keys, per-bottle state, ...)."""
+    return bot_bottle_root() / "db" / HOST_DB_FILENAME
+
+
+__all__ = ["HOST_DB_FILENAME", "bot_bottle_root", "host_db_path"]

bot_bottle/queue_store.py

@@ -7,11 +7,13 @@ import sqlite3
 from pathlib import Path
 
 try:
-    from .supervise_types import Proposal, Response, host_db_path
+    from .supervise_types import Proposal, Response
+    from .paths import 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 supervise_types import Proposal, Response  # type: ignore[import-not-found]  # pylint: disable=import-error,no-name-in-module
+    from paths import 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
 

bot_bottle/store_manager.py

@@ -23,13 +23,10 @@ class StoreManager:
 
     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
+                from .paths 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
+                from paths 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
 

bot_bottle/supervise.py

@@ -41,7 +41,6 @@ try:
     from .supervise_types import (
         ACTION_OPERATOR_EDIT,
         AuditEntry,
-        HOST_DB_FILENAME,
         Proposal,
         Response,
         STATUSES,
@@ -54,13 +53,11 @@ try:
         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,
@@ -73,10 +70,15 @@ except ImportError:
         TOOL_EGRESS_TOKEN_ALLOW,
         TOOL_GITLEAKS_ALLOW,
         TOOL_LIST_EGRESS_ROUTES,
-        bot_bottle_root,
     )
 
 
+try:
+    from .paths import bot_bottle_root
+except ImportError:  # flat imports inside the sidecar bundle
+    from paths import bot_bottle_root  # type: ignore[import-not-found,no-redef]  # pylint: disable=import-error,no-name-in-module
+
+
 SUPERVISE_HOSTNAME = "supervise"
 SUPERVISE_PORT = 9100
 
@@ -106,16 +108,6 @@ 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"
 
@@ -297,8 +289,6 @@ __all__ = [
     "archive_proposal",
     "audit_dir",
     "audit_log_path",
-    "bot_bottle_root",
-    "host_db_path",
     "list_pending_proposals",
     "list_all_pending_proposals",
     "read_audit_entries",

bot_bottle/supervise_types.py

@@ -1,26 +1,19 @@
-"""Shared types and path helpers for supervise stores (PRD 0013).
+"""Shared types 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
+Proposal, Response, and AuditEntry without creating a circular import
-circular import (supervise imports from queue_store/audit_store and
+(supervise imports from queue_store/audit_store and vice-versa).
-vice-versa).
 
-Patching bot_bottle_root on this module propagates through host_db_path
+The path helpers (`bot_bottle_root`, `host_db_path`, `HOST_DB_FILENAME`)
-because host_db_path looks up bot_bottle_root via sys.modules[__name__]
+live in `paths` — import them from there.
-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"
@@ -43,23 +36,6 @@ 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):
@@ -171,7 +147,6 @@ class AuditEntry:
 __all__ = [
     "ACTION_OPERATOR_EDIT",
     "AuditEntry",
-    "HOST_DB_FILENAME",
     "Proposal",
     "Response",
     "STATUSES",
@@ -184,6 +159,4 @@ __all__ = [
     "TOOL_EGRESS_TOKEN_ALLOW",
     "TOOL_GITLEAKS_ALLOW",
     "TOOL_LIST_EGRESS_ROUTES",
-    "bot_bottle_root",
-    "host_db_path",
 ]

docs/prds/0069-firecracker-native-docker-free.md

@@ -1,10 +1,16 @@
 # PRD 0069: Firecracker-native, Docker-free backend
 
-- **Status:** Draft
+- **Status:** Draft (partially superseded)
 - **Author:** Claude
 - **Created:** 2026-07-12
 - **Issue:** #348
 
+> **Superseded in part by [PRD 0070](0070-per-host-orchestrator.md) (#351):**
+> the sidecar-consolidation framing here (Stage 1, per-host sidecar; Stage 4,
+> sidecar-as-VM) is taken over by 0070's per-host orchestrator. This PRD still
+> owns the docker-free **image-building** work — Stage 2 (nix-built fixed
+> images, a dependency of 0070) and Stage 3 (in-VM Dockerfile builder).
+
 ## Summary
 
 Make the Firecracker backend depend on **firecracker + KVM only**, removing

docs/prds/0070-per-host-orchestrator.md

@@ -0,0 +1,433 @@
+# PRD 0070: Per-host orchestrator service
+
+- **Status:** Draft
+- **Author:** Claude
+- **Created:** 2026-07-12
+- **Issue:** #351
+- **Supersedes:** the Stage-1 / Stage-4 sidecar-consolidation framing of
+  PRD 0069 (#348). Depends on 0069's nix-built fixed images (Stage 2) for
+  bootstrapping; 0069 still owns the docker-free image-building work.
+
+## Summary
+
+Replace the **per-bottle sidecar bundle** with a single **persistent,
+per-host orchestrator**: one long-lived service that runs the sidecar
+functions (egress / git-gate / supervise), coordinates with the console,
+and brokers agent launches and teardown. It is **virtualized from the
+start** using each backend's native isolation primitive — a Firecracker
+microVM on the Firecracker backend, an Apple container on macOS, a Docker
+container on the legacy backend — and is fronted by a single
+**backend-agnostic contract**. Per-backend variation lives on
+`BottleBackend`, not in the orchestrator.
+
+## Motivation
+
+Today each bottle spins up its own sidecar bundle (egress mitmproxy +
+git-gate + supervise). That costs:
+
+- **Resources.** N bottles → N heavy bundles booting and idling.
+- **Operational churn.** Per-launch container/VM lifecycle for the
+  sidecars, a control path baked at launch and torn down at exit.
+- **A blurry contract.** "How a bottle talks to its sidecar" is
+  re-implemented per backend instead of being one agreed interface.
+
+A per-host orchestrator collapses the first two and forces the third to
+be made explicit. It's also the component that will own per-host runtime
+**state** (slot leases, the approval queue, the bottle registry) — today
+that's ad-hoc `fcntl`-locked files.
+
+## Security review (read this first)
+
+Consolidation is a real change to the trust model. The goal is to **not
+significantly weaken** the posture; some properties strengthen, some
+weaken, and the weakened ones must be mitigated by design, not hand-waved.
+
+### What gets stronger
+
+- **Build/host isolation of untrusted inputs** (with 0069 Stage 3): user
+  Dockerfiles build in a disposable VM instead of on the host.
+- **One audited privileged surface.** Today the launcher runs as the full
+  host user and needs the Docker socket (root-equivalent). The orchestrator
+  model replaces that with a **thin launch broker** (below) — a small,
+  structured, auditable privileged core instead of a fat socket.
+- **Attribution is enforced, not assumed.** Making source-IP identity a
+  first-class contract invariant (below) means each backend must *prove*
+  it, rather than the sidecar implicitly trusting network position.
+
+### What gets weaker, and the mitigation
+
+1. **Secret concentration.** Per-bottle sidecars isolate secrets at the
+   process boundary — each holds only its bottle's tokens/keys. A host
+   orchestrator concentrates **every bottle's** egress tokens, git deploy
+   keys, and the console credential in one long-lived process. A single
+   attribution bug leaks bottle A's token into bottle B's request — a class
+   of bug that *cannot exist* per-bottle.
+   - *Mitigation:* lean on the enforced source-IP invariant for
+     attribution; keep the most secret-dense, least-shareable service
+     (**git-gate**, per-repo deploy keys, no natural source-IP scoping)
+     **per-bottle** unless there's a compelling reason; scope each secret
+     to the bottle in the state DB so a lookup can't return the wrong
+     bottle's secret by construction (key every secret access by the
+     verified source identity, never by ambient state).
+
+2. **Shared fate.** Orchestrator down = no new launches, and running
+   agents lose egress / git / supervise. Compromise = the whole host's
+   fleet, plus launch authority, plus the console token.
+   - *Mitigation:* the orchestrator is itself confined (its own VM/container
+     with its own fail-closed egress); make it **restartable without killing
+     running agent VMs** (agents keep running; they briefly lose sidecar
+     connectivity until it's back); persist state to a host volume so a
+     restart re-adopts live bottles rather than losing them.
+
+3. **The launch broker is the new privileged core.** We don't eliminate
+   host privilege — we shrink and relocate it. If the broker accepts
+   arbitrary paths/commands, the orchestrator VM can escape through it.
+   - *Mitigation:* the broker takes **structured requests only** — "launch
+     bottle from *this* content-addressed, nix-built rootfs on TAP slot
+     *k*", never "run this argv". It validates against a fixed image set,
+     not caller-supplied paths. It is small enough to audit line-by-line.
+
+4. **The egress proxy now parses every bottle's traffic in one process.**
+   Higher blast radius for a mitmproxy/TLS-bump bug.
+   - *Mitigation:* this is the argument for virtualizing the orchestrator
+     from the start (Stage B, not a host daemon) — the code that TLS-bumps
+     and parses agent traffic and holds every token runs **inside its own
+     confined VM**, not as a host process. If egress sharing's blast radius
+     feels too high, egress can stay per-bottle while supervise (near-zero
+     secrets) goes host-level first.
+
+### The attribution invariant
+
+Source-IP attribution is what makes a shared orchestrator safe: one
+process serves every bottle and tells them apart by source address. The
+*mechanism* is identical everywhere (read source IP → look up bottle); the
+**guarantee that the address can't be forged is a per-backend
+responsibility** and part of the contract:
+
+> **Invariant:** a packet's source address, as seen by the orchestrator,
+> *provably* identifies the originating bottle.
+
+- **Firecracker** — enforced by the `/31` point-to-point TAP + the
+  `bot_bottle_fc` nft table (strongest; already built).
+- **Docker** — the per-bottle `--internal` network + anti-spoof; weaker,
+  must be made explicit.
+- **Apple** — the host-only network.
+
+If a backend can't honor the invariant, source-IP consolidation is not
+safe there and that backend keeps per-bottle sidecars. The invariant is a
+hard precondition, not an aspiration.
+
+**Defense-in-depth — a per-bottle identity token.** On top of the
+network-layer invariant, inject a per-bottle secret token into every
+request the agent makes to the orchestrator (the agent already egresses
+through the sidecar proxy, so this is cheap to add). It gives an
+**application-layer** proof of identity independent of the network layer:
+
+- On **Firecracker** the `/31` + nft already make source IP unspoofable
+  *by construction*, so the token is belt-and-suspenders there — but cheap
+  insurance against a misconfigured invariant.
+- On **weaker backends** (Docker) it is load-bearing, providing attribution
+  that doesn't lean on network anti-spoof.
+
+Requirements: the token is **per-bottle, unguessable, and
+non-cross-leakable** — a bottle can only ever prove it is *itself* (it
+can't learn another bottle's token, given bottle isolation), so a hostile
+agent gains nothing by presenting it. The orchestrator provisions the
+token at launch and the sidecar requires it to attribute + authorize.
+Note this hardens *attribution*, not *secret exposure*: it's app-layer, so
+a compromised orchestrator still sees every token (that's the
+concentration problem, addressed separately under Secret handling).
+
+### Secret handling — a FUTURE pattern (not v1)
+
+> **Status: future / not required for the initial orchestrator.** The
+> initial cut can inject secrets as today; this section records the
+> direction so v1 doesn't paint itself into a corner. Tracked as its own
+> work in **#355** (generic `SecretProvider`).
+
+The residual weakness after all of the above is **long-lived credential
+concentration** — the data-plane proxies must hold every bottle's upstream
+tokens because the agent must never see them. You can't policy-gate the
+component whose job is to *use* all the secrets, so a full RCE of a proxy
+drains its authorized set regardless. Two moves bound this without
+pretending to prevent it:
+
+1. **Vault as a separate trust domain.** The long-lived *roots* live in a
+   distinct process (ideally its own VM) that the byte-parsing data plane
+   never shares memory with. The proxies request secrets from it; the
+   crown jewels are not in the process an agent-facing parser can pop.
+2. **Derive short-lived, scoped creds where the upstream allows it.** The
+   vault holds the root and mints expiring, narrowly-scoped credentials
+   per bottle-start (or per request) — GitHub App installation tokens,
+   OAuth/STS token exchange, forge deploy tokens. A compromise then leaks
+   short-lived material, not permanent keys. For upstreams stuck on static
+   keys the vault passes the value through (only the at-rest / audit /
+   revocation benefits apply, not lifetime reduction) — this residue is
+   accepted and documented, not solved.
+
+Even a plain per-request fetch (no derivation) still buys **at-rest**
+reduction (process memory holds only in-flight secrets), a **detection /
+rate-limit / revocation chokepoint**, and clean **cross-component
+scoping** (an egress RCE can't request git-gate's creds). It does *not*
+prevent abuse of currently-authorized access during the compromise window.
+
+**Mechanism (see #355):** generalize the existing `DeployKeyProvisioner`
+(PRD 0048) into a user-extensible **`SecretProvider`** droppable into the
+manifest anywhere a raw token value is accepted, discovered from
+`~/.bot-bottle/contrib/<name>/secret_provider.py` exactly as user
+`AgentProvider`s are — because maintaining every forge/cloud/OAuth
+provider in-tree is untenable. The orchestrator's vault mints via these
+providers; but the abstraction is shippable independently and today's
+per-bottle sidecars can use it too.
+
+## Design
+
+### The contract (backend-agnostic)
+
+Three surfaces; only one is per-backend.
+
+1. **Control plane (CLI / console → orchestrator)** — an RPC:
+   `launch_bottle`, `teardown_bottle`, `register_policy`,
+   `deregister_bottle`, `supervise_queue`. Fully backend-agnostic. Both the
+   local `cli.py` and the remote console funnel through it, so policy is
+   uniform and `cli.py` becomes a thin client rather than a parallel
+   launcher. **Transport: HTTP** — the most universal/reliable choice on
+   every host (no vsock/unix-socket portability caveats); a local unix
+   socket is a fine optimization, but HTTP is the wire contract.
+2. **Data plane (agent → orchestrator)** — the egress / git / supervise
+   endpoints. Already agnostic today (agents dial `http://sidecar:9099`);
+   only the *address* and *how packets get there* are per-backend.
+3. **Launch / wire (orchestrator → backend)** — the irreducibly
+   backend-specific part; lives on `BottleBackend`.
+
+### One `Orchestrator`, no subclass tree
+
+The orchestrator is a **single concrete class** holding all the
+backend-neutral logic — egress addon, git-gate, supervise, source-IP
+attribution, live-reload control plane, console client. It never branches
+on backend; it *composes* a `BottleBackend`. That composition is what makes
+the contract agnostic: there is nothing backend-specific left in the
+orchestrator to leak.
+
+Rejected alternative: an `Orchestrator` ABC with per-backend
+implementations. The interesting logic (proxies, attribution, control
+plane) is backend-neutral, so three subclasses would triplicate the hard
+part; and a second hierarchy paralleling `BottleBackend` reintroduces the
+same hand-maintained lockstep coupling we just removed from the netpool
+constants (PR #350). Composition over a parallel tree.
+
+### `BottleBackend` absorbs the per-backend variation
+
+A small, cohesive surface — reused for launching agent bottles *and* the
+orchestrator's own unit (the orchestrator is just another native unit):
+
+```
+launch_unit(spec) -> Handle      # agent bottle OR the orchestrator itself
+                                 # (fc microVM / apple ctr / docker ctr)
+wire(unit, endpoint) -> None     # DNAT+forward (fc) | attach shared net (docker/apple)
+endpoint_of(unit) -> Endpoint    # address resolution
+health(unit) -> Status
+```
+
+Plus the **launch broker** — the answer to "a VM/container can't spawn its
+own host-network siblings." The orchestrator can't directly open host
+`/dev/kvm` + a host TAP fd (Firecracker), and a container can't spawn
+siblings without a root-equivalent socket (Docker). So every backend
+exposes a broker the orchestrator calls to launch an agent:
+
+- **Firecracker** — a thin, structured host shim (see security #3). This
+  replaces today's implicit "launcher runs as host user."
+- **Docker** — the socket today (fat, root-equivalent — the thing 0069's
+  Stage 3 removes); a narrower broker later.
+- **Apple** — the `container` CLI/daemon.
+
+**Broker request schema:** human-readable JSON of **static flags + ids
+only** (never free-form paths/argv — that's what keeps it un-coercible
+into arbitrary launches), wrapped as a **signed JWT** so the broker
+verifies *provenance*: the request came from the real orchestrator, not a
+forged one from a compromised co-located component. The orchestrator signs;
+the broker verifies with the orchestrator's public key (provisioned at
+broker install). This is the concrete form of security #3's "structured
+requests only." Same JSON-with-ids + JWT shape for all
+orchestrator↔broker/sidecar communication; cases that don't fit get
+handled as they arise, with this as the default.
+
+If `BottleBackend` bloats, the pressure valve is composition one level
+down: vend a `backend.network()` / `Wiring` collaborator rather than
+piling methods on — the same discipline, recursed.
+
+### State: one SQLite DB, owned by the orchestrator
+
+The orchestrator is the natural owner of per-host **runtime state**:
+
+- pool **slot leases** (which bottle holds slot *i*) — replaces today's
+  `fcntl`-locked files with SQLite transactions;
+- the **supervise approval queue** + remembered approvals;
+- the **live bottle registry** (source IP → bottle → policy/secrets refs),
+  the lookup table the attribution invariant reads.
+
+This is deliberately **not** a "single source of truth for all config."
+Config splits into three tiers with different homes:
+
+| Tier | Example | Home |
+|---|---|---|
+| Build-time constants | pool size, IP base, nft table | flat `.env` (PR #350) — must be readable by Nix eval + root bash, zero runtime |
+| User-authored config | bottle manifests, egress routes, secret refs | declarative files under `~/.bot-bottle/` — trust boundary at `$HOME`, git-trackable, "unknown keys die at load" |
+| Runtime state | slot leases, approvals, registry | one shared **`bot-bottle.db`**, solely owned by the orchestrator |
+
+SQLite is right for the runtime tier (mutable, concurrent, queried) and
+wrong for the other two (Nix can't read it at eval time; it fights the
+declarative manifest trust model). Keep the tiers separate.
+
+**One shared `bot-bottle.db` for all runtime state** (decided in review).
+The registry co-tenants the existing host `bot-bottle.db` — the `DbStore`
+framework already namespaces each store by `schema_key`, so slot
+leases / approvals / registry share one file. One place to query, back up,
+and integrate a console against.
+
+- **Host-resident, for durability.** Re-adoption sweeps the registry after
+  an orchestrator restart, so state *must* outlive the orchestrator
+  instance. The file lives on the host (`bot_bottle_root()/db/bot-bottle.db`);
+  the orchestrator unit reaches it, it doesn't carry it.
+- **Integrity by sole ownership, not mount permissions.** Agents can't
+  touch the DB directly wherever it lives (network-isolated in their
+  bottles). The risk is a *compromised agent-facing data-plane service*
+  (egress/git-gate, which parse hostile bytes) writing the registry and
+  forging attribution. Because it's now one shared file, coarse `ro`/`rw`
+  mount-splitting no longer isolates the registry — so the rule is stronger
+  and simpler: **only the orchestrator (control plane) opens `bot-bottle.db`;
+  the data plane and the console reach state through the control-plane RPC,
+  never a direct file handle.** No agent-facing component gets the file, so
+  none can forge attribution. (This supersedes the earlier `ro`-mount idea.)
+  - *Transitional caveat:* today the per-bottle **supervise sidecar
+    rw-bind-mounts `bot-bottle.db`** to write proposals — exactly the
+    pattern the orchestrator removes (supervise consolidates into the
+    orchestrator; sidecar writes become RPC calls). Until that lands, don't
+    put the attribution registry behind a data-plane-writable mount.
+
+Implementation note for the VM slices: SQLite **WAL** over a guest share
+(virtiofs/9p) is finicky (the `-shm`/`-wal` files need real mmap/locking),
+which is a second reason the DB wants a **host-side owner** the orchestrator
+reaches over the RPC rather than a shared mount into the VM. WAL on the
+shared DB is therefore a deliberate, tested future change — not enabled ad
+hoc. `sqlite3` itself is stdlib, so "the host needs SQLite" is a non-cost.
+
+## Sequencing
+
+Jump straight to the **virtualized** end state (not a host-daemon stepping
+stone): a host daemon's agent→`localhost` transport is throwaway once the
+orchestrator becomes a VM. Decouple the two risks instead:
+
+- **Consolidation risk** (one process, all secrets, attribution, reload)
+  and **packaging/transport risk** (VM-to-VM wiring, the shim) are
+  independent. Develop the orchestrator **service as a plain process
+  dev-harness** first, so the consolidation logic (attribution, reload,
+  secret handling) is proven with fast iteration — *then* wrap that exact
+  service in the VM and solve wiring separately.
+
+### Slices
+
+Built bottom-up as a stack of small PRs off this one (status: **1–5
+implemented** in #352 → #356 → #357 → #358 → #360):
+
+1. **Dev-harness core** ✅ — SQLite registry (runtime state) + fail-closed
+   attribution (source IP + identity token) + the HTTP control plane.
+2. **Launch lifecycle + broker contract** ✅ — `launch_bottle` /
+   `teardown_bottle` + the signed, structured `LaunchBroker` request
+   (HS256 JWT, static ids/flags only, provenance-verified, fail-closed),
+   with a stub broker for the harness and registry rollback on failure.
+3. **Docker launch broker** ✅ — the first real broker: `docker run` / `rm`
+   per bottle, built only from the request's static fields; proves the
+   `BottleBackend` seam on the cheapest backend.
+
+   **The consolidated-sidecar arc** (the core consolidation win — one
+   persistent sidecar shared by all bottles instead of one per bottle):
+
+4. **Sidecar — lifecycle** ✅ — the idempotent-singleton `Sidecar` /
+   `DockerSidecar` (ensure-running is a no-op when already up, stop is
+   idempotent); one container per host.
+5. **Sidecar — build the real bundle image** ✅ — the orchestrator builds
+   the `bot-bottle-sidecars` bundle from `Dockerfile.sidecars` when missing
+   and launches *that* (not a placeholder) as the singleton.
+6. **Sidecar — source-IP-keyed multi-tenant config** — the functional
+   core: the per-bottle CA / egress routes / git-gate keys / policy, today
+   baked per bottle, become **multi-tenant and selected by the verified
+   source IP**, with per-bottle add/remove (live reload) on launch/teardown
+   through the control plane. Until this lands, one shared instance can't
+   serve multiple bottles' distinct policies.
+7. **Sidecar — route agent bottles through it** — wire each agent bottle's
+   egress/git to the shared sidecar (DNAT / network) instead of a per-bottle
+   bundle.
+
+   **Remaining backends** (docker done above; the rest against the proven
+   dev-harness):
+
+8. **Firecracker broker + sidecar** — the real work: the shim + VM-to-VM
+   routing (host forwards `bbfcN` → orchestrator TAP; the nft table grows
+   forward rules where today it drops all non-DNAT egress).
+9. **macOS (Apple container)** — last (container-to-container networking).
+
+Backends land cheapest-first (docker → firecracker → macOS); keep the
+sidecar **service one shared thing** throughout.
+
+## Non-goals
+
+- Removing OCI/Dockerfile support for agent images (0069's concern).
+- A single database for *all* config (see the three-tier table).
+- Changing the per-bottle isolation of agent workloads — only the sidecar
+  is consolidated; agents stay one-VM/container-each.
+
+## Relationship to other work
+
+- **PRD 0069 (#348):** 0070 subsumes its Stage 1 (per-host sidecar) and
+  Stage 4 (sidecar-as-VM). 0069 retains Stage 2 (nix-built fixed images —
+  a **dependency** here: the orchestrator and agent base must be
+  nix-built so the broker launches from a fixed image set and bootstrapping
+  has no chicken-and-egg) and Stage 3 (in-VM Dockerfile builder).
+- **Minimal CI runner (paused):** the Firecracker broker + no host Docker
+  is what lets a dedicated `gitea` runner user drop the root-equivalent
+  `docker` group — it only needs broker-socket access + `kvm`/pool group
+  membership. This work unblocks it.
+- **Generic `SecretProvider` (#355):** the future secret-handling
+  mechanism (see "Secret handling") — generalizes PRD 0048's
+  `DeployKeyProvisioner` into a user-extensible provider that mints
+  short-lived creds. Shippable independently; the orchestrator's vault
+  mints through it.
+- **PR #350 (netpool single-source):** the same "one source per fact,
+  composition over parallel hierarchies" discipline the contract follows.
+
+## Decisions (review 2026-07-13)
+
+- **Egress sharing:** **consolidate egress** (worth it). Treat it as a
+  minimal, hardened attack surface for a malicious agent rather than
+  keeping it per-bottle; pair it with the identity token above and the
+  short-lived-vault-token mitigations (Secret handling / #355).
+- **Control-plane transport:** **HTTP** — most universal/reliable on every
+  host; unix socket is an optional local optimization (see the contract).
+- **Broker request schema:** **signed-JWT JSON, static flags + ids only**
+  (see the launch broker) — provenance + un-coercible by construction.
+- **State re-adoption:** the restart procedure is:
+  1. **Singleton:** a new orchestrator launch requires no pre-existing
+     orchestrator; any found (healthy or not) is fully shut down first.
+  2. Re-adoption **waits for the new orchestrator to be healthy**.
+  3. Once healthy, it discovers all agents needing adoption via **both the
+     SQLite registry and live process/VM inspection** *before serving any
+     other request*. The two-source sweep is what closes the *in-flight
+     launch* race — a launch that started (VM booting / slot claimed) but
+     hadn't committed to SQLite when the old orchestrator died would be
+     invisible to a SQLite-only sweep; process inspection catches it.
+     Launches should also write an **intent record ahead of committing
+     resources** so the sweep can reconcile intent vs. actual.
+
+## Open questions
+
+- **VM-to-VM routing:** per-backend, and in the design it *is*
+  `BottleBackend.wire()` (DNAT+forward for fc, shared-net for
+  docker/apple), not orchestrator logic. **Not a blocker** — resolvable at
+  implementation time (may require a bit of host modification, as the pool
+  setup already does on NixOS); an earlier "sidecars in VMs" spike showed
+  it's feasible.
+- **Live-reload protocol** for per-bottle policy over the HTTP control
+  plane (add/remove routes/keys/proposals without a restart).
+- **Identity-token delivery:** exactly how the per-bottle token is placed
+  where the agent can present it but not swap in another bottle's.

tests/unit/__init__.py

@@ -2,7 +2,7 @@
 
 Isolates ``HOME`` to a throwaway directory for the entire unit suite so
 no test ever reads or writes the real ``~/.bot-bottle`` (state, queue,
-and audit dirs all derive from ``supervise.bot_bottle_root()`` →
+and audit dirs all derive from ``paths.bot_bottle_root()`` →
 ``Path.home()``). Without this, a test that takes a ``flock`` on the
 real audit log can **block indefinitely** when a live bottle's supervise
 sidecar holds that lock — observed as a hung ``coverage run`` at 0% CPU —
@@ -10,8 +10,12 @@ and unisolated tests otherwise pollute the developer's home dir.
 
 Individual tests that need their own ``HOME`` still override
 ``os.environ['HOME']`` and restore it; they now restore to this isolated
-dir rather than the real one, so isolation holds either way. Tests that
+dir rather than the real one, so isolation holds either way.
-patch ``supervise.bot_bottle_root`` directly are unaffected.
+
+Tests that need their own app-data root call ``use_bottle_root`` (below),
+which sets ``BOT_BOTTLE_ROOT`` — the supported override read by
+``paths.bot_bottle_root`` — instead of monkey-patching. One env setting
+covers every module and every flat/package copy of the helper.
 """
 
 from __future__ import annotations
@@ -20,6 +24,9 @@ import atexit
 import os
 import shutil
 import tempfile
+from collections.abc import Callable
+from pathlib import Path
+from unittest import mock
 
 _real_home = os.environ.get("HOME")
 _tmp_home = tempfile.mkdtemp(prefix="bot-bottle-unit-home.")
@@ -35,3 +42,15 @@ def _restore_home() -> None:
 
 
 atexit.register(_restore_home)
+
+
+def use_bottle_root(root: Path) -> Callable[[], None]:
+    """Redirect ``paths.bot_bottle_root()`` to ``root`` by setting
+    ``BOT_BOTTLE_ROOT`` — the supported override. Returns a callable that
+    undoes it (store it as the test's restore hook, or pass to addCleanup).
+
+    Replaces monkey-patching ``paths.bot_bottle_root``; one env var covers
+    every module and the flat/package copies alike (they all read it)."""
+    patcher = mock.patch.dict(os.environ, {"BOT_BOTTLE_ROOT": str(root)})
+    patcher.start()
+    return patcher.stop

tests/unit/test_backend_freezer.py

@@ -7,7 +7,8 @@ import unittest
 from pathlib import Path
 from unittest.mock import patch
 
-from bot_bottle import supervise, bottle_state
+from bot_bottle import bottle_state
+from tests.unit import use_bottle_root
 from bot_bottle.backend import ActiveAgent
 from bot_bottle.backend.freeze import get_freezer
 from bot_bottle.backend.docker.freezer import DockerFreezer
@@ -18,13 +19,7 @@ from bot_bottle.backend.firecracker.freezer import FirecrackerFreezer
 class _FakeHomeMixin:
     def _setup_fake_home(self):
         self._tmp = tempfile.TemporaryDirectory(prefix="freezer-test.")
-        original = supervise.bot_bottle_root
+        self._restore = use_bottle_root(Path(self._tmp.name) / ".bot-bottle")
-
-        def fake_root() -> Path:
-            return Path(self._tmp.name) / ".bot-bottle"
-
-        supervise.bot_bottle_root = fake_root  # type: ignore[assignment]
-        self._restore = lambda: setattr(supervise, "bot_bottle_root", original)
 
     def _teardown_fake_home(self):
         self._restore()

tests/unit/test_backend_prepare.py

@@ -13,7 +13,7 @@ from pathlib import Path
 from unittest.mock import patch
 
 from bot_bottle import bottle_state
-from bot_bottle import supervise
+from tests.unit import use_bottle_root
 from bot_bottle.backend import BottleSpec
 from bot_bottle.backend.docker import DockerBottleBackend
 from bot_bottle.backend.resolve_common import mint_slug
@@ -55,11 +55,10 @@ class _FakeStateMixin:
     def setUp(self) -> None:
         self.tmp = tempfile.TemporaryDirectory(prefix="backend-prepare.")
         self.root = Path(self.tmp.name) / ".bot-bottle"
-        self.original_root = supervise.bot_bottle_root
+        self._restore = use_bottle_root(self.root)
-        supervise.bot_bottle_root = lambda: self.root  # type: ignore[assignment]
 
     def tearDown(self) -> None:
-        supervise.bot_bottle_root = self.original_root  # type: ignore[assignment]
+        self._restore()
         self.tmp.cleanup()
 
 

tests/unit/test_backend_workspace.py

@@ -13,7 +13,7 @@ from pathlib import Path
 from unittest.mock import MagicMock, call, patch
 
 from bot_bottle import bottle_state
-from bot_bottle import supervise
+from tests.unit import use_bottle_root
 from bot_bottle.backend import Bottle, BottleSpec, ExecResult
 from bot_bottle.backend.docker import DockerBottleBackend
 from bot_bottle.backend.firecracker import FirecrackerBottleBackend
@@ -55,11 +55,10 @@ class _FakeStateMixin:
         self.tmp_dir = tempfile.TemporaryDirectory(prefix="backend-workspace.")
         self.tmp = Path(self.tmp_dir.name)
         self.root = self.tmp / ".bot-bottle"
-        self.original_root = supervise.bot_bottle_root
+        self._restore = use_bottle_root(self.root)
-        supervise.bot_bottle_root = lambda: self.root  # type: ignore[assignment]
 
     def tearDown(self) -> None:
-        supervise.bot_bottle_root = self.original_root  # type: ignore[assignment]
+        self._restore()
         self.tmp_dir.cleanup()
 
 

tests/unit/test_bottle_state.py

@@ -6,7 +6,7 @@ import tempfile
 import unittest
 from pathlib import Path
 
-from bot_bottle import supervise
+from tests.unit import use_bottle_root
 from bot_bottle import bottle_state
 from bot_bottle.bottle_state import (
     BottleMetadata,
@@ -18,13 +18,7 @@ from bot_bottle.bottle_state import (
 class _FakeHomeMixin:
     def _setup_fake_home(self):
         self._tmp = tempfile.TemporaryDirectory(prefix="bottle-state-test.")
-        original = supervise.bot_bottle_root
+        self._restore = use_bottle_root(Path(self._tmp.name) / ".bot-bottle")
-
-        def fake_root() -> Path:
-            return Path(self._tmp.name) / ".bot-bottle"
-
-        supervise.bot_bottle_root = fake_root  # type: ignore[assignment]
-        self._restore = lambda: setattr(supervise, "bot_bottle_root", original)
 
     def _teardown_fake_home(self):
         self._restore()

tests/unit/test_cli_commit.py

@@ -8,7 +8,7 @@ from pathlib import Path
 from unittest.mock import MagicMock, patch
 
 from bot_bottle.cli.commit import cmd_commit
-from bot_bottle import supervise
+from tests.unit import use_bottle_root
 from bot_bottle import bottle_state
 from bot_bottle.backend.freeze import CommitCancelled
 
@@ -16,13 +16,7 @@ from bot_bottle.backend.freeze import CommitCancelled
 class _FakeHomeMixin:
     def _setup_fake_home(self):
         self._tmp = tempfile.TemporaryDirectory(prefix="cli-commit-test.")
-        original = supervise.bot_bottle_root
+        self._restore = use_bottle_root(Path(self._tmp.name) / ".bot-bottle")
-
-        def fake_root() -> Path:
-            return Path(self._tmp.name) / ".bot-bottle"
-
-        supervise.bot_bottle_root = fake_root  # type: ignore[assignment]
-        self._restore = lambda: setattr(supervise, "bot_bottle_root", original)
 
     def _teardown_fake_home(self):
         self._restore()

tests/unit/test_cli_start_settle.py

@@ -8,7 +8,7 @@ import tempfile
 import unittest
 from pathlib import Path
 
-from bot_bottle import supervise
+from tests.unit import use_bottle_root
 from bot_bottle import bottle_state
 from bot_bottle.cli import start as start_mod
 
@@ -16,15 +16,10 @@ from bot_bottle.cli import start as start_mod
 class _FakeHomeMixin:
     def _setup_fake_home(self):
         self._tmp = tempfile.TemporaryDirectory(prefix="cli-start-settle.")
-        self._original = supervise.bot_bottle_root
+        self._restore = use_bottle_root(Path(self._tmp.name) / ".bot-bottle")
-
-        def fake_root() -> Path:
-            return Path(self._tmp.name) / ".bot-bottle"
-
-        supervise.bot_bottle_root = fake_root  # type: ignore[assignment]
 
     def _teardown_fake_home(self):
-        supervise.bot_bottle_root = self._original  # type: ignore[assignment]
+        self._restore()
         self._tmp.cleanup()
 
 

tests/unit/test_docker_cleanup.py

@@ -15,7 +15,7 @@ import tempfile
 import unittest
 from pathlib import Path
 
-from bot_bottle import supervise
+from tests.unit import use_bottle_root
 from bot_bottle import bottle_state
 from bot_bottle.backend.docker.cleanup import _list_orphan_state_dirs
 
@@ -23,13 +23,7 @@ from bot_bottle.backend.docker.cleanup import _list_orphan_state_dirs
 class _FakeHomeMixin:
     def _setup_fake_home(self) -> None:
         self._tmp = tempfile.TemporaryDirectory(prefix="docker-cleanup-test.")
-        original = supervise.bot_bottle_root
+        self._restore = use_bottle_root(Path(self._tmp.name) / ".bot-bottle")
-
-        def fake_root() -> Path:
-            return Path(self._tmp.name) / ".bot-bottle"
-
-        supervise.bot_bottle_root = fake_root  # type: ignore[assignment]
-        self._restore = lambda: setattr(supervise, "bot_bottle_root", original)
 
     def _teardown_fake_home(self) -> None:
         self._restore()

tests/unit/test_docker_enumerate_active.py

@@ -23,7 +23,7 @@ import tempfile
 import unittest
 from pathlib import Path
 
-from bot_bottle import supervise
+from tests.unit import use_bottle_root
 from bot_bottle import bottle_state
 from bot_bottle.backend.docker import enumerate as _enumerate
 
@@ -75,13 +75,7 @@ class TestParseServicesByProject(unittest.TestCase):
 class _FakeHomeMixin:
     def _setup_fake_home(self) -> None:
         self._tmp = tempfile.TemporaryDirectory(prefix="enum-active.")
-        original = supervise.bot_bottle_root
+        self._restore_home = use_bottle_root(Path(self._tmp.name) / ".bot-bottle")
-
-        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)
 
     def _teardown_fake_home(self) -> None:
         self._restore_home()

tests/unit/test_egress_apply.py

@@ -8,7 +8,7 @@ from pathlib import Path
 from types import SimpleNamespace
 from unittest.mock import patch
 
-from bot_bottle import supervise
+from tests.unit import use_bottle_root
 from bot_bottle.backend.egress_apply import EgressApplyError
 from bot_bottle.backend.docker.egress_apply import applicator
 
@@ -67,14 +67,8 @@ class TestValidateRoutesContent(unittest.TestCase):
 class TestApplyRoutesChange(unittest.TestCase):
     def setUp(self):
         self._tmp = tempfile.TemporaryDirectory(prefix="egress-apply-test.")
-        original = supervise.bot_bottle_root
-
-        def fake_root() -> Path:
-            return Path(self._tmp.name) / ".bot-bottle"
-
-        supervise.bot_bottle_root = fake_root  # type: ignore[assignment]
-        self.addCleanup(lambda: setattr(supervise, "bot_bottle_root", original))
         self.addCleanup(self._tmp.cleanup)
+        self.addCleanup(use_bottle_root(Path(self._tmp.name) / ".bot-bottle"))
 
     def test_writes_live_routes_and_signals_reload(self):
         calls: list[list[str]] = []

tests/unit/test_orchestrator_control_plane.py

@@ -0,0 +1,150 @@
+"""Unit tests for the orchestrator HTTP control plane (PRD 0070).
+
+Mostly exercises the pure `dispatch()` (socket-free, like the supervise
+server tests), plus one real-socket round-trip to prove the handler wiring.
+"""
+
+from __future__ import annotations
+
+import json
+import tempfile
+import threading
+import unittest
+import urllib.request
+from pathlib import Path
+
+from bot_bottle.orchestrator.control_plane import dispatch, make_server
+from bot_bottle.orchestrator.registry import RegistryStore
+
+
+def _body(obj: object) -> bytes:
+    return json.dumps(obj).encode()
+
+
+class TestDispatch(unittest.TestCase):
+    def setUp(self) -> None:
+        self._tmp = tempfile.TemporaryDirectory()
+        self.store = RegistryStore(Path(self._tmp.name) / "r.db")
+        self.store.migrate()
+
+    def tearDown(self) -> None:
+        self._tmp.cleanup()
+
+    def test_health(self) -> None:
+        status, payload = dispatch(self.store, "GET", "/health", b"")
+        self.assertEqual(200, status)
+        self.assertEqual("ok", payload["status"])
+
+    def test_register_returns_id_and_token(self) -> None:
+        status, payload = dispatch(
+            self.store, "POST", "/bottles", _body({"source_ip": "10.243.0.1"})
+        )
+        self.assertEqual(201, status)
+        self.assertTrue(payload["bottle_id"])
+        self.assertTrue(payload["identity_token"])
+
+    def test_register_requires_source_ip(self) -> None:
+        status, _ = dispatch(self.store, "POST", "/bottles", _body({}))
+        self.assertEqual(400, status)
+
+    def test_register_rejects_bad_json(self) -> None:
+        status, _ = dispatch(self.store, "POST", "/bottles", b"{not json")
+        self.assertEqual(400, status)
+
+    def test_list_redacts_identity_token(self) -> None:
+        dispatch(self.store, "POST", "/bottles", _body({"source_ip": "10.243.0.1"}))
+        status, payload = dispatch(self.store, "GET", "/bottles", b"")
+        self.assertEqual(200, status)
+        bottles = payload["bottles"]
+        assert isinstance(bottles, list)
+        self.assertEqual(1, len(bottles))
+        first = bottles[0]
+        assert isinstance(first, dict)
+        self.assertNotIn("identity_token", first)
+        self.assertEqual("10.243.0.1", first["source_ip"])
+
+    def test_attribute_ok_and_forbidden(self) -> None:
+        _, reg = dispatch(
+            self.store, "POST", "/bottles", _body({"source_ip": "10.243.0.5"})
+        )
+        token = reg["identity_token"]
+        ok_status, ok = dispatch(
+            self.store, "POST", "/attribute",
+            _body({"source_ip": "10.243.0.5", "identity_token": token}),
+        )
+        self.assertEqual(200, ok_status)
+        self.assertEqual(reg["bottle_id"], ok["bottle_id"])
+
+        bad_status, _ = dispatch(
+            self.store, "POST", "/attribute",
+            _body({"source_ip": "10.243.0.5", "identity_token": "nope"}),
+        )
+        self.assertEqual(403, bad_status)
+
+    def test_attribute_requires_both_fields(self) -> None:
+        status, _ = dispatch(
+            self.store, "POST", "/attribute", _body({"source_ip": "10.243.0.5"})
+        )
+        self.assertEqual(400, status)
+
+    def test_delete(self) -> None:
+        _, reg = dispatch(
+            self.store, "POST", "/bottles", _body({"source_ip": "10.243.0.1"})
+        )
+        bid = reg["bottle_id"]
+        status, payload = dispatch(self.store, "DELETE", f"/bottles/{bid}", b"")
+        self.assertEqual(200, status)
+        self.assertEqual(True, payload["deregistered"])
+        self.assertEqual([], self.store.all())
+
+    def test_delete_missing_404(self) -> None:
+        status, _ = dispatch(self.store, "DELETE", "/bottles/ghost", b"")
+        self.assertEqual(404, status)
+
+    def test_unknown_route_404(self) -> None:
+        status, _ = dispatch(self.store, "GET", "/nope", b"")
+        self.assertEqual(404, status)
+
+    def test_trailing_slash_normalized(self) -> None:
+        status, _ = dispatch(self.store, "GET", "/health/", b"")
+        self.assertEqual(200, status)
+
+
+class TestServerRoundTrip(unittest.TestCase):
+    def test_http_register_health_attribute(self) -> None:
+        tmp = tempfile.TemporaryDirectory()
+        self.addCleanup(tmp.cleanup)
+        store = RegistryStore(Path(tmp.name) / "r.db")
+        store.migrate()
+        server = make_server(store, "127.0.0.1", 0)
+        self.addCleanup(server.server_close)
+        thread = threading.Thread(target=server.serve_forever, daemon=True)
+        thread.start()
+        self.addCleanup(server.shutdown)
+
+        host, port = server.server_address[0], server.server_address[1]
+        base = f"http://{host}:{port}"
+
+        reg = json.load(urllib.request.urlopen(
+            urllib.request.Request(
+                f"{base}/bottles", data=_body({"source_ip": "10.243.0.7"}),
+                method="POST", headers={"Content-Type": "application/json"},
+            ), timeout=5,
+        ))
+        self.assertTrue(reg["bottle_id"])
+
+        health = json.load(urllib.request.urlopen(f"{base}/health", timeout=5))
+        self.assertEqual("ok", health["status"])
+
+        attr = json.load(urllib.request.urlopen(
+            urllib.request.Request(
+                f"{base}/attribute",
+                data=_body({"source_ip": "10.243.0.7", "identity_token": reg["identity_token"]}),
+                method="POST", headers={"Content-Type": "application/json"},
+            ), timeout=5,
+        ))
+        self.assertEqual(reg["bottle_id"], attr["bottle_id"])
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/unit/test_orchestrator_registry.py

@@ -0,0 +1,102 @@
+"""Unit tests for the orchestrator bottle registry + attribution (PRD 0070)."""
+
+from __future__ import annotations
+
+import tempfile
+import unittest
+from pathlib import Path
+
+from bot_bottle.orchestrator.registry import (
+    BottleRecord,
+    RegistryStore,
+    new_identity_token,
+)
+
+
+class TestRegistryStore(unittest.TestCase):
+    def setUp(self) -> None:
+        self._tmp = tempfile.TemporaryDirectory()
+        self.db = Path(self._tmp.name) / "registry.db"
+        self.store = RegistryStore(self.db)
+        self.store.migrate()
+
+    def tearDown(self) -> None:
+        self._tmp.cleanup()
+
+    def test_register_mints_id_and_token(self) -> None:
+        rec = self.store.register("10.243.0.1")
+        self.assertTrue(rec.bottle_id)
+        self.assertTrue(rec.identity_token)
+        self.assertEqual("active", rec.state)
+        self.assertEqual(rec, self.store.get(rec.bottle_id))
+
+    def test_identity_tokens_are_unique(self) -> None:
+        tokens = {new_identity_token() for _ in range(200)}
+        self.assertEqual(200, len(tokens))
+        a = self.store.register("10.243.0.1")
+        b = self.store.register("10.243.0.3")
+        self.assertNotEqual(a.identity_token, b.identity_token)
+
+    def test_all_and_deregister(self) -> None:
+        a = self.store.register("10.243.0.1")
+        b = self.store.register("10.243.0.3")
+        self.assertEqual({a.bottle_id, b.bottle_id}, {r.bottle_id for r in self.store.all()})
+        self.assertTrue(self.store.deregister(a.bottle_id))
+        self.assertEqual([b.bottle_id], [r.bottle_id for r in self.store.all()])
+
+    def test_deregister_missing_is_false(self) -> None:
+        self.assertFalse(self.store.deregister("nope"))
+
+    def test_explicit_id_replaces(self) -> None:
+        self.store.register("10.243.0.1", bottle_id="fixed", metadata="a")
+        self.store.register("10.243.0.9", bottle_id="fixed", metadata="b")
+        rec = self.store.get("fixed")
+        assert rec is not None
+        self.assertEqual("10.243.0.9", rec.source_ip)
+        self.assertEqual("b", rec.metadata)
+        self.assertEqual(1, len(self.store.all()))
+
+    def test_attribute_success_requires_ip_and_token(self) -> None:
+        rec = self.store.register("10.243.0.1")
+        got = self.store.attribute("10.243.0.1", rec.identity_token)
+        assert got is not None
+        self.assertEqual(rec.bottle_id, got.bottle_id)
+
+    def test_attribute_wrong_token_denied(self) -> None:
+        self.store.register("10.243.0.1")
+        self.assertIsNone(self.store.attribute("10.243.0.1", "wrong-token"))
+
+    def test_attribute_unknown_ip_denied(self) -> None:
+        rec = self.store.register("10.243.0.1")
+        self.assertIsNone(self.store.attribute("10.243.9.9", rec.identity_token))
+
+    def test_attribute_empty_token_denied(self) -> None:
+        self.store.register("10.243.0.1")
+        self.assertIsNone(self.store.attribute("10.243.0.1", ""))
+
+    def test_attribute_ambiguous_ip_denied(self) -> None:
+        # Two active bottles on one source IP is a misconfiguration — deny
+        # rather than guess (fail-closed), even with a valid token.
+        a = self.store.register("10.243.0.1", bottle_id="a")
+        self.store.register("10.243.0.1", bottle_id="b")
+        self.assertIsNone(self.store.attribute("10.243.0.1", a.identity_token))
+
+    def test_state_persists_across_reopen(self) -> None:
+        rec = self.store.register("10.243.0.1")
+        reopened = RegistryStore(self.db)
+        got = reopened.get(rec.bottle_id)
+        assert got is not None
+        self.assertEqual(rec, got)
+        # Attribution works against the reopened (durable) store too.
+        self.assertIsNotNone(reopened.attribute("10.243.0.1", rec.identity_token))
+
+    def test_redacted_hides_token(self) -> None:
+        rec = BottleRecord(
+            bottle_id="x", source_ip="10.243.0.1", identity_token="secret"
+        )
+        self.assertNotIn("identity_token", rec.redacted())
+        self.assertEqual("10.243.0.1", rec.redacted()["source_ip"])
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/unit/test_supervise.py

@@ -8,7 +8,8 @@ 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.paths import host_db_path
+from tests.unit import use_bottle_root
 from bot_bottle.audit_store import AuditStore
 from bot_bottle.queue_store import QueueStore
 from bot_bottle.supervise import (
@@ -21,7 +22,6 @@ from bot_bottle.supervise import (
     TOOL_EGRESS_ALLOW,
     TOOL_GITLEAKS_ALLOW,
     archive_proposal,
-    host_db_path,
     list_pending_proposals,
     read_audit_entries,
     read_proposal,
@@ -123,20 +123,7 @@ class TestQueueIO(unittest.TestCase):
         self._tmp.cleanup()
 
     def _patch_home(self, fake_home: Path):
-        original_sv = supervise.bot_bottle_root
+        return use_bottle_root(fake_home / ".bot-bottle")
-        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()
@@ -240,20 +227,7 @@ class TestAuditLog(unittest.TestCase):
         self._tmp.cleanup()
 
     def _patch_home(self, fake_home: Path):
-        original_sv = supervise.bot_bottle_root
+        return use_bottle_root(fake_home / ".bot-bottle")
-        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_then_read_single_entry(self):
         e = AuditEntry(
@@ -400,20 +374,7 @@ class TestSupervisePrepare(unittest.TestCase):
         self._tmp.cleanup()
 
     def _patch_home(self, fake_home: Path):
-        original_sv = supervise.bot_bottle_root
+        return use_bottle_root(fake_home / ".bot-bottle")
-        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_prepare_creates_queue(self):
         plan = _StubSupervise().prepare("dev", self.stage_dir)

tests/unit/test_supervise_cli.py

@@ -12,7 +12,7 @@ from pathlib import Path
 from unittest.mock import patch
 
 from bot_bottle import supervise
-from bot_bottle import supervise_types as sv_types
+from tests.unit import use_bottle_root
 from bot_bottle.audit_store import AuditStore
 from bot_bottle.cli import supervise as supervise_cli
 from bot_bottle.queue_store import QueueStore
@@ -51,24 +51,11 @@ def _proposal(slug: str = "dev", tool: str = TOOL_EGRESS_ALLOW) -> Proposal:
 
 
 class _FakeHomeMixin:
-    """Patch supervise.bot_bottle_root to a temp dir for the test."""
+    """Point bot_bottle_root at a temp dir (via BOT_BOTTLE_ROOT) for the test."""
 
     def _setup_fake_home(self):
         self._tmp = tempfile.TemporaryDirectory(prefix="supervise-test.")
-        original_sv = supervise.bot_bottle_root
+        self._restore_home = use_bottle_root(Path(self._tmp.name) / ".bot-bottle")
-        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]
-        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()
 

tests/unit/test_supervise_cli_crash_logging.py

@@ -11,12 +11,13 @@ from __future__ import annotations
 
 import contextlib
 import io
+import os
 import tempfile
 import unittest
 from pathlib import Path
 from unittest import mock
 
-from bot_bottle import supervise
+from tests.unit import use_bottle_root
 from bot_bottle.cli import supervise as supervise_cli
 from bot_bottle.log import Die, die
 
@@ -39,18 +40,16 @@ class TestDieCarriesMessage(unittest.TestCase):
 
 
 class _FakeHomeMixin:
-    """Point supervise.bot_bottle_root (what _write_crash_log resolves
+    """Point bot_bottle_root (what _write_crash_log resolves through) at a
-    through) at a temp dir so the crash log doesn't touch the real
+    temp dir so the crash log doesn't touch the real ~/.bot-bottle."""
-    ~/.bot-bottle."""
 
     def _setup_fake_home(self):
         self._tmp = tempfile.TemporaryDirectory(prefix="supervise-crash-test.")
-        self._orig_root = supervise.bot_bottle_root
         self._root = Path(self._tmp.name) / ".bot-bottle"
-        supervise.bot_bottle_root = lambda: self._root  # type: ignore[assignment]
+        self._restore_root = use_bottle_root(self._root)
 
     def _teardown_fake_home(self):
-        supervise.bot_bottle_root = self._orig_root  # type: ignore[assignment]
+        self._restore_root()
         self._tmp.cleanup()
 
 
@@ -127,11 +126,11 @@ class TestWriteCrashLog(_FakeHomeMixin, unittest.TestCase):
         # OSError and the helper must fall back to a tempfile.
         bad = Path(self._tmp.name) / "not-a-dir"
         bad.write_text("x")
-        supervise.bot_bottle_root = lambda: bad  # type: ignore[assignment]
+        with mock.patch.dict(os.environ, {"BOT_BOTTLE_ROOT": str(bad)}):
-        try:
+            try:
-            raise RuntimeError("explode2")
+                raise RuntimeError("explode2")
-        except RuntimeError as e:
+            except RuntimeError as e:
-            path = supervise_cli._write_crash_log(e)
+                path = supervise_cli._write_crash_log(e)
         self.assertTrue(path.exists())
         self.assertIn("explode2", path.read_text())
 

tests/unit/test_supervise_edge.py

@@ -12,6 +12,7 @@ from unittest.mock import patch
 
 from bot_bottle import supervise
 from bot_bottle.audit_store import AuditStore
+from bot_bottle.paths import bot_bottle_root
 from bot_bottle.queue_store import QueueStore
 from bot_bottle.supervise import (
     AuditEntry,
@@ -39,7 +40,7 @@ def _proposal() -> Proposal:
 
 class TestPathHelpers(unittest.TestCase):
     def test_bot_bottle_root(self) -> None:
-        self.assertTrue(str(supervise.bot_bottle_root()).endswith(".bot-bottle"))
+        self.assertTrue(str(bot_bottle_root()).endswith(".bot-bottle"))
 
 
 class TestReadMalformed(unittest.TestCase):

tests/unit/test_supervise_server.py

@@ -10,6 +10,8 @@ import unittest
 from pathlib import Path
 from unittest.mock import patch
 
+from tests.unit import use_bottle_root
+
 
 # The server module loads `supervise` via same-directory import inside
 # the container (Dockerfile.supervise WORKDIRs into /app). For tests
@@ -19,10 +21,8 @@ sys.path.insert(0, str(Path(__file__).resolve().parent.parent.parent / "bot_bott
 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,
@@ -279,23 +279,7 @@ class TestHandleToolsCall(unittest.TestCase):
         self._tmp.cleanup()
 
     def _patch_home(self, fake_home: Path):
-        original_sv = _sv.bot_bottle_root
+        return use_bottle_root(fake_home / ".bot-bottle")
-        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