feat(mcp): Phase 5 — write-time near-duplicate gate (update-over-create)

#755 Phase 5. create_note / create_task now BLOCK a near-duplicate instead of
silently inserting: they return {"duplicate": true, "existing_id", message}
pointing at the record to UPDATE. Fights store bloat and stale competing copies
that semantic search (RAG) would otherwise resurface for reconciliation. A
force=true override creates anyway for genuinely-distinct records.

- services/dedup.py: find_duplicate_note — two signals, scoped to owner + same
  project + same kind: (1) normalized-title exact match (cheap, always); (2)
  semantic cosine ≥ 0.90 but ONLY when body ≥ 200 chars (short/title-only
  embeddings false-positive — the pre-pivot lesson). Project-less (orphan)
  records compare only to other orphans on BOTH signals (orphan_only on the
  semantic call) — they're not matched across every project.
- Gate wired into the MCP create_note/create_task tools (the LLM write path)
  with force override; _INSTRUCTIONS documents the duplicate response + force.
- Opt-in by design: the service helper is only called from the interactive
  create tools. Internal/programmatic creates (recurrence spawn, imports) go
  straight through services.create_note and are NOT gated — a recurring task
  spawning its next same-titled instance must not be blocked.
- Scope v1: MCP tools only. REST/web (human CRUD, needs a UI affordance) and
  create_rule (not a RAG surface; _INSTRUCTIONS already steer it) are follow-ups.
- tests: dedup service (title/semantic/body-gate/type-filter) + tool gate
  (blocks, force bypasses) for notes and tasks.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

tests/test_mcp_tool_notes.py

@@ -25,6 +25,32 @@ def _fake_note(**overrides) -> MagicMock:
     return note
 
 
+@pytest.mark.asyncio
+async def test_create_note_blocked_by_duplicate_gate():
+    from scribe.services.dedup import DuplicateMatch
+    dup = DuplicateMatch(id=88, title="Embeddings notes", similarity=0.94, reason="semantic")
+    create_mock = AsyncMock()
+    with patch("scribe.mcp.tools.notes.dedup_svc.find_duplicate_note",
+               AsyncMock(return_value=dup)), \
+         patch("scribe.mcp.tools.notes.notes_svc.create_note", create_mock):
+        out = await create_note(title="embeddings", body="notes about embeddings")
+    assert out["duplicate"] is True
+    assert out["existing_id"] == 88
+    assert out["match"] == "semantic"
+    create_mock.assert_not_called()
+
+
+@pytest.mark.asyncio
+async def test_create_note_force_bypasses_duplicate_gate():
+    find_mock = AsyncMock()
+    with patch("scribe.mcp.tools.notes.dedup_svc.find_duplicate_note", find_mock), \
+         patch("scribe.mcp.tools.notes.notes_svc.create_note",
+               AsyncMock(return_value=_fake_note(id=3))):
+        out = await create_note(title="dup", force=True)
+    assert out["id"] == 3
+    find_mock.assert_not_called()
+
+
 @pytest.mark.asyncio
 async def test_list_notes_repackages_tuple_into_dict():
     rows = [_fake_note(id=1), _fake_note(id=2)]

tests/test_mcp_tool_tasks.py

@@ -113,6 +113,33 @@ async def test_create_task_passes_status():
     assert mock.call_args.kwargs["status"] == "todo"
 
 
+@pytest.mark.asyncio
+async def test_create_task_blocked_by_duplicate_gate():
+    """A near-duplicate blocks creation and returns the existing id (no insert)."""
+    from scribe.services.dedup import DuplicateMatch
+    dup = DuplicateMatch(id=42, title="Set up CI", similarity=1.0, reason="title")
+    create_mock = AsyncMock()
+    with patch("scribe.mcp.tools.tasks.dedup_svc.find_duplicate_note",
+               AsyncMock(return_value=dup)), \
+         patch("scribe.mcp.tools.tasks.notes_svc.create_note", create_mock):
+        out = await create_task(title="set up ci")
+    assert out["duplicate"] is True
+    assert out["existing_id"] == 42
+    create_mock.assert_not_called()  # nothing was created
+
+
+@pytest.mark.asyncio
+async def test_create_task_force_bypasses_duplicate_gate():
+    """force=true skips the gate entirely and creates."""
+    find_mock = AsyncMock()
+    with patch("scribe.mcp.tools.tasks.dedup_svc.find_duplicate_note", find_mock), \
+         patch("scribe.mcp.tools.tasks.notes_svc.create_note",
+               AsyncMock(return_value=_fake_task(id=9))):
+        out = await create_task(title="dup", force=True)
+    assert out["id"] == 9
+    find_mock.assert_not_called()  # gate not even consulted
+
+
 @pytest.mark.asyncio
 async def test_create_task_rejects_retired_plan_kind():
     """kind=plan is hard-retired — plans are milestones (start_planning)."""

tests/test_services_dedup.py

@@ -0,0 +1,84 @@
+"""Unit tests for the write-time near-duplicate gate (services/dedup.py)."""
+from unittest.mock import AsyncMock, MagicMock, patch
+
+import pytest
+
+from scribe.services.dedup import DuplicateMatch, duplicate_response, find_duplicate_note
+
+
+def _session_returning(note):
+    """A mocked async_session() whose single execute() yields `note` (or None)."""
+    s = AsyncMock()
+    s.__aenter__ = AsyncMock(return_value=s)
+    s.__aexit__ = AsyncMock(return_value=False)
+    result = MagicMock()
+    result.scalars.return_value.first.return_value = note
+    s.execute = AsyncMock(return_value=result)
+    return s
+
+
+def _fake_note(id=1, title="T", note_type="note"):
+    n = MagicMock()
+    n.id, n.title, n.note_type = id, title, note_type
+    return n
+
+
+@pytest.mark.asyncio
+async def test_title_exact_match_returns_title_duplicate():
+    note = _fake_note(id=10, title="Setup CI")
+    with patch("scribe.services.dedup.async_session",
+               return_value=_session_returning(note)):
+        # whitespace/case differences are normalized away
+        dup = await find_duplicate_note(7, "  setup   ci ", project_id=2, is_task=True)
+    assert dup is not None
+    assert dup.id == 10
+    assert dup.reason == "title"
+    assert dup.similarity == 1.0
+
+
+@pytest.mark.asyncio
+async def test_short_body_skips_semantic_check():
+    sem = AsyncMock()
+    with patch("scribe.services.dedup.async_session",
+               return_value=_session_returning(None)), \
+         patch("scribe.services.dedup.embeddings_svc.semantic_search_notes", sem):
+        dup = await find_duplicate_note(7, "Unique", body="too short", project_id=2)
+    assert dup is None
+    sem.assert_not_called()  # body under _MIN_BODY_FOR_SEMANTIC
+
+
+@pytest.mark.asyncio
+async def test_semantic_match_when_body_substantial():
+    hit = _fake_note(id=20, title="Existing", note_type="note")
+    sem = AsyncMock(return_value=[(0.93, hit)])
+    with patch("scribe.services.dedup.async_session",
+               return_value=_session_returning(None)), \
+         patch("scribe.services.dedup.embeddings_svc.semantic_search_notes", sem):
+        dup = await find_duplicate_note(
+            7, "Title", body="x" * 250, project_id=2, is_task=False, note_type="note",
+        )
+    assert dup is not None
+    assert dup.id == 20
+    assert dup.reason == "semantic"
+    assert dup.similarity == 0.93
+
+
+@pytest.mark.asyncio
+async def test_semantic_match_of_other_note_type_is_ignored():
+    other = _fake_note(id=21, title="X", note_type="process")
+    sem = AsyncMock(return_value=[(0.97, other)])
+    with patch("scribe.services.dedup.async_session",
+               return_value=_session_returning(None)), \
+         patch("scribe.services.dedup.embeddings_svc.semantic_search_notes", sem):
+        dup = await find_duplicate_note(7, "Title", body="x" * 250, note_type="note")
+    assert dup is None  # type mismatch must not block
+
+
+def test_duplicate_response_shape():
+    dm = DuplicateMatch(id=5, title="Foo", similarity=1.0, reason="title")
+    r = duplicate_response(dm, "task")
+    assert r["duplicate"] is True
+    assert r["existing_id"] == 5
+    assert r["match"] == "title"
+    assert "force=true" in r["message"]
+    assert "update_task" in r["message"]