refactor(supervise): make Supervisor a service class the orchestrator calls

Turn the loose queue/audit functions in orchestrator/supervisor/queue.py into
methods on a concrete Supervisor service. The Orchestrator now owns an
injectable `self._supervisor` and calls `self._supervisor.write_proposal(...)`
etc., instead of module-level free functions — the supervise dependency is
explicit and mockable, and a Supervisor can be scoped to a `db_path` (defaults
to the host DB) so tests can point it at a temp database.

  - Supervisor (in the package __init__) drops the ABC and gains write_proposal,
    read_proposal, list_pending_proposals, list_all_pending_proposals,
    write_response, read_response, archive_all_proposals, write_audit_entry,
    read_audit_entries, plus the launch-time prepare().
  - render_diff / sha256_hex are pure and stateless, so they move to
    orchestrator/supervisor/util.py (re-exported from the facade) rather than
    becoming methods.
  - queue.py is deleted; service.py + the supervise tests call through a
    Supervisor instance.

Full unit suite green (2243).

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

tests/unit/test_orchestrator_control_plane.py

@@ -29,7 +29,7 @@ from bot_bottle.orchestrator.supervisor import (
     Proposal,
     TOOL_EGRESS_ALLOW,
     sha256_hex,
-    write_proposal,
+    Supervisor,
 )
 
 
@@ -421,7 +421,7 @@ class TestDispatchSupervise(unittest.TestCase):
         p = Proposal.new(
             bottle_slug=slug, tool=TOOL_EGRESS_ALLOW, proposed_file=proposed,
             justification="need it", current_file_hash=sha256_hex(proposed))
-        write_proposal(p)
+        Supervisor().write_proposal(p)
         return p.id
 
     def test_list_pending(self) -> None:

tests/unit/test_orchestrator_service.py

@@ -20,9 +20,8 @@ from bot_bottle.orchestrator.supervisor import (
     Proposal,
     STATUS_APPROVED,
     TOOL_EGRESS_ALLOW,
-    read_response,
     sha256_hex,
-    write_proposal,
+    Supervisor,
 )
 
 
@@ -185,7 +184,7 @@ class TestOrchestratorSupervise(unittest.TestCase):
         p = Proposal.new(
             bottle_slug=slug, tool=TOOL_EGRESS_ALLOW, proposed_file=proposed,
             justification="need it", current_file_hash=sha256_hex(proposed))
-        write_proposal(p)
+        Supervisor().write_proposal(p)
         return p.id
 
     def test_pending_lists_queued_proposal(self) -> None:
@@ -208,7 +207,7 @@ class TestOrchestratorSupervise(unittest.TestCase):
         assert rec is not None
         self.assertEqual(new_routes, rec.policy)
         # response written -> agent unblocks, proposal no longer pending
-        self.assertEqual(STATUS_APPROVED, read_response("demo", pid).status)
+        self.assertEqual(STATUS_APPROVED, Supervisor().read_response("demo", pid).status)
         self.assertEqual([], self.orch.supervise_pending())
 
     def test_pending_carries_human_label(self) -> None:
@@ -261,7 +260,7 @@ class TestOrchestratorSupervise(unittest.TestCase):
         rec = self.store.get(bottle_id)
         assert rec is not None
         self.assertEqual("routes:\n  - host: existing.com\n", rec.policy)
-        self.assertEqual("rejected", read_response("demo", pid).status)
+        self.assertEqual("rejected", Supervisor().read_response("demo", pid).status)
 
     def test_unknown_proposal_is_error(self) -> None:
         ok, err = self.orch.supervise_respond(

tests/unit/test_supervise.py

@@ -17,19 +17,15 @@ from bot_bottle.orchestrator.supervisor import (
     STATUS_APPROVED,
     STATUS_MODIFIED,
     STATUS_REJECTED,
+    Supervisor,
     TOOL_EGRESS_ALLOW,
     TOOL_GITLEAKS_ALLOW,
-    list_pending_proposals,
-    read_audit_entries,
-    read_proposal,
-    read_response,
     render_diff,
     sha256_hex,
-    write_audit_entry,
-    write_proposal,
-    write_response,
 )
 
+_SV = Supervisor()
+
 
 FIXED_TS = datetime(2026, 5, 25, 12, 0, 0, tzinfo=timezone.utc)
 
@@ -123,26 +119,26 @@ class TestQueueIO(unittest.TestCase):
 
     def test_write_and_read_proposal(self):
         p = _proposal()
-        path = write_proposal(p)
+        path = _SV.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.slug, p.id)
+        loaded = _SV.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(a)
-        write_proposal(b)
-        write_response(self.slug, Response(
+        _SV.write_proposal(a)
+        _SV.write_proposal(b)
+        _SV.write_response(self.slug, Response(
             proposal_id=a.id, status=STATUS_APPROVED, notes="",
         ))
-        pending = list_pending_proposals(self.slug)
+        pending = _SV.list_pending_proposals(self.slug)
         self.assertEqual([b.id], [p.id for p in pending])
 
     def test_list_pending_returns_empty_for_missing_slug(self):
-        self.assertEqual([], list_pending_proposals("nope"))
+        self.assertEqual([], _SV.list_pending_proposals("nope"))
 
     def test_list_pending_sorted_by_arrival(self):
         # Fabricate two with explicit timestamps.
@@ -159,15 +155,15 @@ class TestQueueIO(unittest.TestCase):
             now=datetime(2026, 5, 25, 14, 0, 0, tzinfo=timezone.utc),
         )
         # Write in reverse order.
-        write_proposal(b)
-        write_proposal(a)
-        ordered = list_pending_proposals(self.slug)
+        _SV.write_proposal(b)
+        _SV.write_proposal(a)
+        ordered = _SV.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.slug, r)
-        self.assertEqual(r, read_response(self.slug, "xyz"))
+        _SV.write_response(self.slug, r)
+        self.assertEqual(r, _SV.read_response(self.slug, "xyz"))
 
 
 class TestAuditLog(unittest.TestCase):
@@ -193,15 +189,15 @@ class TestAuditLog(unittest.TestCase):
             justification="agent needed gh-api token",
             diff="--- before\n+++ after\n",
         )
-        path = write_audit_entry(e)
+        path = _SV.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")
+        loaded = _SV.read_audit_entries("cred-proxy", "dev")
         self.assertEqual([e], loaded)
 
     def test_appends_one_line_per_entry(self):
         for i in range(3):
-            write_audit_entry(AuditEntry(
+            _SV.write_audit_entry(AuditEntry(
                 timestamp=f"2026-05-25T12:00:0{i}+00:00",
                 bottle_slug="dev",
                 component="egress",
@@ -210,7 +206,7 @@ class TestAuditLog(unittest.TestCase):
                 justification="",
                 diff="",
             ))
-        entries = read_audit_entries("egress", "dev")
+        entries = _SV.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",
@@ -219,7 +215,7 @@ class TestAuditLog(unittest.TestCase):
         )
 
     def test_separate_logs_per_component_slug(self):
-        write_audit_entry(AuditEntry(
+        _SV.write_audit_entry(AuditEntry(
             timestamp="t",
             bottle_slug="dev",
             component="cred-proxy",
@@ -228,7 +224,7 @@ class TestAuditLog(unittest.TestCase):
             justification="",
             diff="",
         ))
-        write_audit_entry(AuditEntry(
+        _SV.write_audit_entry(AuditEntry(
             timestamp="t",
             bottle_slug="dev",
             component="egress",
@@ -237,7 +233,7 @@ class TestAuditLog(unittest.TestCase):
             justification="",
             diff="",
         ))
-        write_audit_entry(AuditEntry(
+        _SV.write_audit_entry(AuditEntry(
             timestamp="t",
             bottle_slug="other",
             component="cred-proxy",
@@ -246,12 +242,12 @@ class TestAuditLog(unittest.TestCase):
             justification="",
             diff="",
         ))
-        self.assertEqual(1, len(read_audit_entries("cred-proxy", "dev")))
-        self.assertEqual(1, len(read_audit_entries("egress", "dev")))
-        self.assertEqual(1, len(read_audit_entries("cred-proxy", "other")))
+        self.assertEqual(1, len(_SV.read_audit_entries("cred-proxy", "dev")))
+        self.assertEqual(1, len(_SV.read_audit_entries("egress", "dev")))
+        self.assertEqual(1, len(_SV.read_audit_entries("cred-proxy", "other")))
 
     def test_read_audit_entries_missing_log_returns_empty(self):
-        self.assertEqual([], read_audit_entries("cred-proxy", "no-such-bottle"))
+        self.assertEqual([], _SV.read_audit_entries("cred-proxy", "no-such-bottle"))
 
 
 class TestDiffAndHash(unittest.TestCase):

tests/unit/test_supervise_edge.py

@@ -17,14 +17,12 @@ from bot_bottle.orchestrator.supervisor import (
     AuditEntry,
     Proposal,
     STATUS_APPROVED,
+    Supervisor,
     TOOL_EGRESS_ALLOW,
-    list_pending_proposals,
-    read_audit_entries,
-    read_proposal,
-    read_response,
-    write_audit_entry,
 )
 
+_SV = Supervisor()
+
 
 def _proposal() -> Proposal:
     return Proposal.new(
@@ -47,21 +45,21 @@ class TestReadMalformed(unittest.TestCase):
             with patch.dict("os.environ", {"HOME": d}):
                 QueueStore("slug").migrate()
                 with self.assertRaises(FileNotFoundError):
-                    read_proposal("slug", "p")
+                    _SV.read_proposal("slug", "p")
 
     def test_read_response_missing_row(self) -> None:
         with tempfile.TemporaryDirectory() as d:
             with patch.dict("os.environ", {"HOME": d}):
                 QueueStore("slug").migrate()
                 with self.assertRaises(FileNotFoundError):
-                    read_response("slug", "p")
+                    _SV.read_response("slug", "p")
 
     def test_list_pending_reads_db_only(self) -> None:
         with tempfile.TemporaryDirectory() as d:
             with patch.dict("os.environ", {"HOME": d}):
                 QueueStore("slug").migrate()
-                supervise.write_proposal(_proposal())
-                pending = list_pending_proposals("slug")
+                _SV.write_proposal(_proposal())
+                pending = _SV.list_pending_proposals("slug")
             self.assertEqual(1, len(pending))
             self.assertEqual("slug", pending[0].bottle_slug)
 
@@ -70,26 +68,26 @@ class TestReadMalformed(unittest.TestCase):
             with patch.dict("os.environ", {"HOME": d}):
                 QueueStore("slug").migrate()
                 p = _proposal()
-                supervise.write_proposal(p)
-                supervise.write_response("slug", supervise.Response(
+                _SV.write_proposal(p)
+                _SV.write_response("slug", supervise.Response(
                     proposal_id=p.id,
                     status=STATUS_APPROVED,
                     notes="",
                 ))
-                self.assertEqual([], list_pending_proposals("slug"))
+                self.assertEqual([], _SV.list_pending_proposals("slug"))
 
 
 class TestReadAuditEntries(unittest.TestCase):
     def test_missing_log_returns_empty(self) -> None:
         with tempfile.TemporaryDirectory() as home, \
                 patch.dict("os.environ", {"HOME": home}):
-            self.assertEqual([], read_audit_entries("egress", "nope"))
+            self.assertEqual([], _SV.read_audit_entries("egress", "nope"))
 
     def test_reads_entries_from_db(self) -> None:
         with tempfile.TemporaryDirectory() as home, \
                 patch.dict("os.environ", {"HOME": home}):
             AuditStore().migrate()
-            write_audit_entry(AuditEntry(
+            _SV.write_audit_entry(AuditEntry(
                 timestamp="t",
                 bottle_slug="slug",
                 component="egress",
@@ -98,7 +96,7 @@ class TestReadAuditEntries(unittest.TestCase):
                 justification="",
                 diff="",
             ))
-            write_audit_entry(AuditEntry(
+            _SV.write_audit_entry(AuditEntry(
                 timestamp="t",
                 bottle_slug="other",
                 component="egress",
@@ -107,7 +105,7 @@ class TestReadAuditEntries(unittest.TestCase):
                 justification="",
                 diff="",
             ))
-            entries = read_audit_entries("egress", "slug")
+            entries = _SV.read_audit_entries("egress", "slug")
             self.assertEqual(1, len(entries))
             self.assertEqual("approve", entries[0].operator_action)
 

tests/unit/test_supervise_server.py

@@ -50,6 +50,11 @@ from bot_bottle.gateway.supervise_server import (
     validate_proposed_file,
 )
 
+# The orchestrator's supervise service, backing the fake resolver's queue reads
+# and writes (the fake stands in for the RPC; state still goes through the real
+# queue store).
+_SV = _sv.Supervisor()
+
 # Fixed caller identity for the handler tests. The control plane attributes by
 # (source_ip, identity_token); the fake resolver ignores them and answers for a
 # fixed bottle, since attribution itself is covered by the orchestrator tests.
@@ -89,7 +94,7 @@ class _FakeSuperviseResolver:
             bottle_slug=self.bottle_id, tool=tool, proposed_file=proposed_file,
             justification=justification, current_file_hash=_sv.sha256_hex(proposed_file),
         )
-        _sv.write_proposal(proposal)
+        _SV.write_proposal(proposal)
         return proposal.id
 
     def poll_supervise(
@@ -101,10 +106,10 @@ class _FakeSuperviseResolver:
         if self.bottle_id is None or self.poll_none:
             return None
         try:
-            response = _sv.read_response(self.bottle_id, proposal_id)
+            response = _SV.read_response(self.bottle_id, proposal_id)
         except FileNotFoundError:
             try:
-                _sv.read_proposal(self.bottle_id, proposal_id)
+                _SV.read_proposal(self.bottle_id, proposal_id)
             except FileNotFoundError:
                 return {"status": _sv.POLL_STATUS_UNKNOWN}
             return {"status": _sv.POLL_STATUS_PENDING}
@@ -377,10 +382,10 @@ class TestHandleToolsCall(unittest.TestCase):
         matching response — the operator half, out of band."""
         def runner():
             for _ in range(200):
-                pending = _sv.list_pending_proposals("dev")
+                pending = _SV.list_pending_proposals("dev")
                 if pending:
                     p = pending[0]
-                    _sv.write_response("dev", _sv.Response(
+                    _SV.write_response("dev", _sv.Response(
                         proposal_id=p.id, status=status, notes=notes,
                     ))
                     return
@@ -487,7 +492,7 @@ class TestHandleToolsCall(unittest.TestCase):
             responder.join()
         # A decided proposal drops off the operator's pending list (a response
         # row exists) — poll itself no longer archives (issue #469 review).
-        self.assertEqual([], _sv.list_pending_proposals("dev"))
+        self.assertEqual([], _SV.list_pending_proposals("dev"))
 
     def test_pending_response_times_out_without_archive(self):
         result = _tools_call(
@@ -505,7 +510,7 @@ 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("dev")))
+        self.assertEqual(1, len(_SV.list_pending_proposals("dev")))
 
     _ALLOW: dict[str, object] = {
         "name": _sv.TOOL_EGRESS_ALLOW,
@@ -747,7 +752,7 @@ class TestNonBlockingSupervise(unittest.TestCase):
             justification="need example.com",
             current_file_hash=_sv.sha256_hex(self._ROUTES),
         )
-        _sv.write_proposal(p)
+        _SV.write_proposal(p)
         return p
 
     # --- pending response carries the id ---
@@ -771,7 +776,7 @@ class TestNonBlockingSupervise(unittest.TestCase):
         self.assertFalse(result["isError"])  # type: ignore[index]
         text = result["content"][0]["text"]  # type: ignore[index]
         self.assertIn("status: pending", text)
-        pending = _sv.list_pending_proposals("dev")
+        pending = _SV.list_pending_proposals("dev")
         self.assertEqual(1, len(pending))  # still queued, not archived
         self.assertIn(pending[0].id, text)  # agent got the id to poll
 
@@ -779,7 +784,7 @@ class TestNonBlockingSupervise(unittest.TestCase):
 
     def test_check_returns_approved_idempotently(self):
         p = self._seed_proposal()
-        _sv.write_response("dev", _sv.Response(proposal_id=p.id, status=_sv.STATUS_APPROVED, notes="ok"))
+        _SV.write_response("dev", _sv.Response(proposal_id=p.id, status=_sv.STATUS_APPROVED, notes="ok"))
         result = _check(self.resolver, {"arguments": {"proposal_id": p.id}})
         self.assertFalse(result["isError"])
         text = result["content"][0]["text"]  # type: ignore[index]
@@ -792,7 +797,7 @@ class TestNonBlockingSupervise(unittest.TestCase):
 
     def test_check_rejected_sets_isError(self):
         p = self._seed_proposal()
-        _sv.write_response("dev", _sv.Response(proposal_id=p.id, status=_sv.STATUS_REJECTED, notes="no"))
+        _SV.write_response("dev", _sv.Response(proposal_id=p.id, status=_sv.STATUS_REJECTED, notes="no"))
         result = _check(self.resolver, {"arguments": {"proposal_id": p.id}})
         self.assertTrue(result["isError"])
         self.assertIn("status: rejected", result["content"][0]["text"])  # type: ignore[index]
@@ -804,7 +809,7 @@ class TestNonBlockingSupervise(unittest.TestCase):
         text = result["content"][0]["text"]  # type: ignore[index]
         self.assertIn("status: pending", text)
         self.assertIn(p.id, text)
-        self.assertEqual(1, len(_sv.list_pending_proposals("dev")))  # not archived
+        self.assertEqual(1, len(_SV.list_pending_proposals("dev")))  # not archived
 
     def test_check_unknown_id_is_error(self):
         result = _check(self.resolver, {"arguments": {"proposal_id": "no-such-proposal"}})
@@ -848,15 +853,15 @@ class TestNonBlockingSupervise(unittest.TestCase):
             },
             ServerConfig(response_timeout_seconds=0.05),
         )
-        pid = _sv.list_pending_proposals("dev")[0].id
+        pid = _SV.list_pending_proposals("dev")[0].id
         self.assertIn(pid, result["content"][0]["text"])  # type: ignore[index]
         # 2. operator decides out-of-band
-        _sv.write_response("dev", _sv.Response(proposal_id=pid, status=_sv.STATUS_APPROVED, notes="ok"))
+        _SV.write_response("dev", _sv.Response(proposal_id=pid, status=_sv.STATUS_APPROVED, notes="ok"))
         # 3. agent resumes by polling — no re-proposing
         poll = _check(self.resolver, {"arguments": {"proposal_id": pid}})
         self.assertFalse(poll["isError"])
         self.assertIn("status: approved", poll["content"][0]["text"])  # type: ignore[index]
-        self.assertEqual([], _sv.list_pending_proposals("dev"))  # resolved (response exists)
+        self.assertEqual([], _SV.list_pending_proposals("dev"))  # resolved (response exists)
 
 
 if __name__ == "__main__":