threadkeeper 0.20.0__tar.gz → 0.22.0__tar.gz

{threadkeeper-0.20.0 → threadkeeper-0.22.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: threadkeeper
-Version: 0.20.0
+Version: 0.22.0
 Summary: Multi-agent shared brain across Claude Code/Desktop, Codex, Antigravity CLI, Copilot, and VS Code. Cross-session memory, self-improving skill loops, inter-agent signaling — one local MCP server.
 Author: thread-keeper contributors
 License: MIT
@@ -39,7 +39,7 @@ Provides-Extra: encrypted
 Requires-Dist: keyring>=25.0; extra == "encrypted"
 Requires-Dist: sqlcipher3-wheels>=0.5.7; extra == "encrypted"
 Provides-Extra: dev
-Requires-Dist: pytest>=9.1.1; extra == "dev"
+Requires-Dist: pytest<9.2,>=9.1.1; extra == "dev"
 Requires-Dist: pytest-cov>=7.1.0; extra == "dev"
 Requires-Dist: pytest-forked>=1.6; extra == "dev"
 Requires-Dist: pytest-xdist>=3.5; extra == "dev"
@@ -972,8 +972,12 @@ recovery snapshot right before its first mutating evaluator.
 its token cost); the evaluator then judges currency from local evidence only.
 A batch that holds no skills never gets a researcher: the fixed research topics
 only concern skills, so its evaluator starts at once.
-Each report path is then
-authorized in the existing parent-authored `curator_pass` event;
+Curator pass and snapshot values are explicit, per-child launch data rather
+than temporary server-process environment. Ordinary foreground and daemon
+spawns strip stale inherited values. The private launcher binds each task,
+child identity, pass, batch, phase, role, origin, and exact
+report/research destination before launch; a watchdog retry restores only that
+recorded data.
 `curator_report_write` records its SHA-256 in `curator_report_provenance`.
 This makes the report directory an untrusted transport: a stray or forged
 `REPORT-*.md` file cannot acquire the provenance needed by the applier.
@@ -1055,6 +1059,18 @@ within an 8-second budget per call; a call that runs out falls back to the
 lexical ranking and the next call resumes. Each line shows the lesson's summary. Without a query a session
 would only ever see the newest few lessons, whatever the task.
 
+**Lesson index.** Agents rarely search lessons on their own, so every lesson
+write also refreshes `lesson-index.md` next to `lessons.md`
+(`~/.threadkeeper/lesson-index.md` by default), a sorted list of all lesson
+titles (about 35 000 characters for 550 lessons). The managed
+instructions import it (`@~/.threadkeeper/lesson-index.md`), so a CLI that
+supports imports, such as Claude Code, loads every title with its instructions;
+other CLIs are told to read the file at session start. Each title is the rule,
+and the agent opens the ones that apply with `lesson_get`. On 160 real user
+prompts an agent that saw only this index opened 65-70% of the lessons judged
+relevant, and 51-59% of its picks were relevant. Injecting the lessons whose
+embedding is closest to the prompt reached 25% recall at 12% precision.
+
 Lesson access is tracked the same way skill access is: `lesson_list` increments
 `lesson_usage.view_count` for displayed rows and `lesson_get` increments
 `lesson_usage.use_count` for the returned lesson. Only foreground reads count:
@@ -1448,6 +1464,13 @@ claim and its evidence but closes the old valid-time interval at the new claim's
 slice; `dialectic_review(as_of=...)` and
 `dialectic_synthesis(include_history=True)` expose past validity intervals.
 
+**Durable-preference drift pilot.** Run
+`python -m threadkeeper.dialectic_drift_eval --json` to replay the fixed,
+anonymized durable-change / task-exception / ambiguity / poisoning corpus. It
+reports supersession precision and recall, false durable-flip and clarification
+rates, poisoning containment, cited memory IDs, and counterfactuals. The pilot
+is read-only and does not change the live supersession policy.
+
 **Source-based evidence discount.** Each evidence row's effective weight
 is `base_weight × discount(WRITE_ORIGIN)`. Foreground (direct user / human
 signal) = 1.0. shadow_review / background_review / candidate_review /
@@ -2251,6 +2274,21 @@ It reports these headline groups over a fixed ground-truth set:
 - **retrieval latency** — mean / p50 / p95 / max wall-clock milliseconds. With
   `--semantic`, the backend is reported as `hybrid`, because dense candidates
   augment rather than replace FTS.
+- **forgetting lifecycle** — the bundled mutation fixture writes a fact,
+  corrects it twice, supersedes its dialectic claim, consolidates duplicate
+  verbatim evidence, forgets the old source, and reopens the store in a new
+  process. It reports a separate score for thread notes, verbatim memory,
+  dialectic claims, and lessons or skills.
+
+- **operational recovery comparison** — a second fully synthetic fixture
+  compares dialogue-only evidence with a bounded action/outcome record. Its
+  fixed thresholds require useful recovery links, zero forbidden-string leaks,
+  bounded storage relative to the dialogue baseline, and zero false causal
+  links. The record shape contains only normalized action names, opaque
+  objective/source references, coarse outcomes, error fingerprints, ordering,
+  recovery links, authority, provenance, and expiry. It has no raw arguments,
+  raw output, transcript text, paths, or secret field.
+
 - **outcome classification** — retrieval failure, reasoning failure,
   stale-memory use, removed-memory use, and privacy-policy violation.
 
@@ -2261,6 +2299,17 @@ unavailable when the optional semantic extra is not installed. `--strict`
 requires the versioned fixture thresholds and zero safety violations, making
 the same command suitable for retrieval/model migration gates.
 
+Action/outcome capture is an **offline-only prototype**. The runtime has no
+live producer or export route. Evaluation records expire after 30 days, and
+the retention pass removes them; `tk-forget --selector-type objective <ref>
+--apply` removes an objective's records and invalidates any linked lesson.
+Records sourced from a forgotten dialog/session/thread are removed by the
+same targeted-erasure path. The existing candidate reviewer may see only the
+bounded recovery candidate; a lesson must keep its
+`action_outcome_candidate:<id>` provenance so invalidation hides it. Live use
+requires a later explicit opt-in plus a passing offline comparison; this
+release provides neither.
+
 For DB concurrency, run the reproducible local gate:
 
 ```bash
@@ -2286,6 +2335,13 @@ retrieved context, or replayed answer is written to stdout, artifacts, or the
 repository. See [docs/ARCHITECTURE.md](docs/ARCHITECTURE.md) for the corpus
 contract and migration-gate interpretation.
 
+The lifecycle score gives current evidence `+1` and subtracts `0.5` for every
+obsolete or invalidated fact that the same retrieval path returns. A current
+fact alongside stale evidence therefore does not receive full credit. The
+post-forgetting and post-restart stages must score `1.0` in both deterministic
+FTS and hybrid variants; the JSON report names the lifecycle stage and memory
+class for any release-gate failure.
+
 ## Evaluating learning-loop decision quality
 
 `verify_ingest` answers *"did we capture the data?"*. The decision-quality

threadkeeper-0.20.0/threadkeeper.egg-info/PKG-INFO → threadkeeper-0.22.0/README.md

@@ -1,50 +1,3 @@
-Metadata-Version: 2.4
-Name: threadkeeper
-Version: 0.20.0
-Summary: Multi-agent shared brain across Claude Code/Desktop, Codex, Antigravity CLI, Copilot, and VS Code. Cross-session memory, self-improving skill loops, inter-agent signaling — one local MCP server.
-Author: thread-keeper contributors
-License: MIT
-Project-URL: Homepage, https://github.com/po4erk91/thread-keeper
-Project-URL: Repository, https://github.com/po4erk91/thread-keeper
-Project-URL: Issues, https://github.com/po4erk91/thread-keeper/issues
-Project-URL: Documentation, https://github.com/po4erk91/thread-keeper#readme
-Project-URL: Changelog, https://github.com/po4erk91/thread-keeper/releases
-Keywords: mcp,model-context-protocol,claude,codex,antigravity,copilot,memory,agents,self-improving,skills
-Classifier: Development Status :: 4 - Beta
-Classifier: License :: OSI Approved :: MIT License
-Classifier: Operating System :: MacOS
-Classifier: Operating System :: POSIX :: Linux
-Classifier: Programming Language :: Python :: 3.11
-Classifier: Programming Language :: Python :: 3.12
-Classifier: Programming Language :: Python :: 3.13
-Classifier: Topic :: Software Development :: Libraries :: Python Modules
-Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Python: >=3.11
-Description-Content-Type: text/markdown
-License-File: LICENSE
-Requires-Dist: mcp<3,>=2.3.0
-Requires-Dist: pydantic>=2.13.4
-Requires-Dist: pydantic-settings>=2.14.2
-Requires-Dist: pyyaml>=6.0.3
-Provides-Extra: semantic
-Requires-Dist: fastembed<0.9,>=0.8; extra == "semantic"
-Requires-Dist: huggingface-hub>=0.20; extra == "semantic"
-Requires-Dist: numpy>=1.24.0; extra == "semantic"
-Requires-Dist: sqlite-vec>=0.1.9; extra == "semantic"
-Provides-Extra: semantic-st
-Requires-Dist: sentence-transformers>=2.2.0; extra == "semantic-st"
-Requires-Dist: numpy>=1.24.0; extra == "semantic-st"
-Requires-Dist: sqlite-vec>=0.1.9; extra == "semantic-st"
-Provides-Extra: encrypted
-Requires-Dist: keyring>=25.0; extra == "encrypted"
-Requires-Dist: sqlcipher3-wheels>=0.5.7; extra == "encrypted"
-Provides-Extra: dev
-Requires-Dist: pytest>=9.1.1; extra == "dev"
-Requires-Dist: pytest-cov>=7.1.0; extra == "dev"
-Requires-Dist: pytest-forked>=1.6; extra == "dev"
-Requires-Dist: pytest-xdist>=3.5; extra == "dev"
-Dynamic: license-file
-
 # thread-keeper
 
 [![tests](https://github.com/po4erk91/thread-keeper/actions/workflows/test.yml/badge.svg)](https://github.com/po4erk91/thread-keeper/actions/workflows/test.yml)
@@ -972,8 +925,12 @@ recovery snapshot right before its first mutating evaluator.
 its token cost); the evaluator then judges currency from local evidence only.
 A batch that holds no skills never gets a researcher: the fixed research topics
 only concern skills, so its evaluator starts at once.
-Each report path is then
-authorized in the existing parent-authored `curator_pass` event;
+Curator pass and snapshot values are explicit, per-child launch data rather
+than temporary server-process environment. Ordinary foreground and daemon
+spawns strip stale inherited values. The private launcher binds each task,
+child identity, pass, batch, phase, role, origin, and exact
+report/research destination before launch; a watchdog retry restores only that
+recorded data.
 `curator_report_write` records its SHA-256 in `curator_report_provenance`.
 This makes the report directory an untrusted transport: a stray or forged
 `REPORT-*.md` file cannot acquire the provenance needed by the applier.
@@ -1055,6 +1012,18 @@ within an 8-second budget per call; a call that runs out falls back to the
 lexical ranking and the next call resumes. Each line shows the lesson's summary. Without a query a session
 would only ever see the newest few lessons, whatever the task.
 
+**Lesson index.** Agents rarely search lessons on their own, so every lesson
+write also refreshes `lesson-index.md` next to `lessons.md`
+(`~/.threadkeeper/lesson-index.md` by default), a sorted list of all lesson
+titles (about 35 000 characters for 550 lessons). The managed
+instructions import it (`@~/.threadkeeper/lesson-index.md`), so a CLI that
+supports imports, such as Claude Code, loads every title with its instructions;
+other CLIs are told to read the file at session start. Each title is the rule,
+and the agent opens the ones that apply with `lesson_get`. On 160 real user
+prompts an agent that saw only this index opened 65-70% of the lessons judged
+relevant, and 51-59% of its picks were relevant. Injecting the lessons whose
+embedding is closest to the prompt reached 25% recall at 12% precision.
+
 Lesson access is tracked the same way skill access is: `lesson_list` increments
 `lesson_usage.view_count` for displayed rows and `lesson_get` increments
 `lesson_usage.use_count` for the returned lesson. Only foreground reads count:
@@ -1448,6 +1417,13 @@ claim and its evidence but closes the old valid-time interval at the new claim's
 slice; `dialectic_review(as_of=...)` and
 `dialectic_synthesis(include_history=True)` expose past validity intervals.
 
+**Durable-preference drift pilot.** Run
+`python -m threadkeeper.dialectic_drift_eval --json` to replay the fixed,
+anonymized durable-change / task-exception / ambiguity / poisoning corpus. It
+reports supersession precision and recall, false durable-flip and clarification
+rates, poisoning containment, cited memory IDs, and counterfactuals. The pilot
+is read-only and does not change the live supersession policy.
+
 **Source-based evidence discount.** Each evidence row's effective weight
 is `base_weight × discount(WRITE_ORIGIN)`. Foreground (direct user / human
 signal) = 1.0. shadow_review / background_review / candidate_review /
@@ -2251,6 +2227,21 @@ It reports these headline groups over a fixed ground-truth set:
 - **retrieval latency** — mean / p50 / p95 / max wall-clock milliseconds. With
   `--semantic`, the backend is reported as `hybrid`, because dense candidates
   augment rather than replace FTS.
+- **forgetting lifecycle** — the bundled mutation fixture writes a fact,
+  corrects it twice, supersedes its dialectic claim, consolidates duplicate
+  verbatim evidence, forgets the old source, and reopens the store in a new
+  process. It reports a separate score for thread notes, verbatim memory,
+  dialectic claims, and lessons or skills.
+
+- **operational recovery comparison** — a second fully synthetic fixture
+  compares dialogue-only evidence with a bounded action/outcome record. Its
+  fixed thresholds require useful recovery links, zero forbidden-string leaks,
+  bounded storage relative to the dialogue baseline, and zero false causal
+  links. The record shape contains only normalized action names, opaque
+  objective/source references, coarse outcomes, error fingerprints, ordering,
+  recovery links, authority, provenance, and expiry. It has no raw arguments,
+  raw output, transcript text, paths, or secret field.
+
 - **outcome classification** — retrieval failure, reasoning failure,
   stale-memory use, removed-memory use, and privacy-policy violation.
 
@@ -2261,6 +2252,17 @@ unavailable when the optional semantic extra is not installed. `--strict`
 requires the versioned fixture thresholds and zero safety violations, making
 the same command suitable for retrieval/model migration gates.
 
+Action/outcome capture is an **offline-only prototype**. The runtime has no
+live producer or export route. Evaluation records expire after 30 days, and
+the retention pass removes them; `tk-forget --selector-type objective <ref>
+--apply` removes an objective's records and invalidates any linked lesson.
+Records sourced from a forgotten dialog/session/thread are removed by the
+same targeted-erasure path. The existing candidate reviewer may see only the
+bounded recovery candidate; a lesson must keep its
+`action_outcome_candidate:<id>` provenance so invalidation hides it. Live use
+requires a later explicit opt-in plus a passing offline comparison; this
+release provides neither.
+
 For DB concurrency, run the reproducible local gate:
 
 ```bash
@@ -2286,6 +2288,13 @@ retrieved context, or replayed answer is written to stdout, artifacts, or the
 repository. See [docs/ARCHITECTURE.md](docs/ARCHITECTURE.md) for the corpus
 contract and migration-gate interpretation.
 
+The lifecycle score gives current evidence `+1` and subtracts `0.5` for every
+obsolete or invalidated fact that the same retrieval path returns. A current
+fact alongside stale evidence therefore does not receive full credit. The
+post-forgetting and post-restart stages must score `1.0` in both deterministic
+FTS and hybrid variants; the JSON report names the lifecycle stage and memory
+class for any release-gate failure.
+
 ## Evaluating learning-loop decision quality
 
 `verify_ingest` answers *"did we capture the data?"*. The decision-quality

{threadkeeper-0.20.0 → threadkeeper-0.22.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "threadkeeper"
-version = "0.20.0"
+version = "0.22.0"
 description = "Multi-agent shared brain across Claude Code/Desktop, Codex, Antigravity CLI, Copilot, and VS Code. Cross-session memory, self-improving skill loops, inter-agent signaling — one local MCP server."
 requires-python = ">=3.11"
 authors = [{ name = "thread-keeper contributors" }]
@@ -70,7 +70,9 @@ encrypted = [
 # not compose with --forked (collides on the xdist worker tmp dir) and a few
 # tests still race under it. See CONTRIBUTING → Running tests.
 dev = [
-    "pytest>=9.1.1",
+    # pytest 9.2.0 breaks pytest-forked 1.7.5: every forked test exits with
+    # status 3 before reporting. Lift the cap once pytest-forked supports it.
+    "pytest>=9.1.1,<9.2",
     "pytest-cov>=7.1.0",
     "pytest-forked>=1.6",
     "pytest-xdist>=3.5",

threadkeeper-0.22.0/tests/test_action_outcomes.py

@@ -0,0 +1,266 @@
+"""Offline action/outcome memory stays bounded, deletable, and non-live."""
+from __future__ import annotations
+
+import json
+import time
+
+import pytest
+
+
+def _record_pair(conn, action_outcomes, *, objective="objective-42", source_id="dialog-42"):
+    failed = action_outcomes.record_evaluation_event(
+        conn,
+        event_id="ao-failed",
+        objective_ref=objective,
+        action_name="github.pr.create",
+        outcome="failure",
+        error_fingerprint="http.422.validation",
+        sequence=1,
+        source_kind="dialog",
+        source_id=source_id,
+        authority="observed",
+        principal="fixture-runner",
+        channel="offline-evaluation",
+        created_at=100,
+    )
+    recovered = action_outcomes.record_evaluation_event(
+        conn,
+        event_id="ao-recovered",
+        objective_ref=objective,
+        action_name="github.pr.create",
+        outcome="success",
+        sequence=2,
+        recovered_from=failed,
+        source_kind="dialog",
+        source_id=source_id,
+        authority="observed",
+        principal="fixture-runner",
+        channel="offline-evaluation",
+        created_at=101,
+    )
+    return failed, recovered
+
+
+def test_schema_and_validation_exclude_raw_payloads(fresh_mp):
+    from threadkeeper import action_outcomes
+
+    conn = fresh_mp["db"].get_db()
+    columns = {
+        row["name"]
+        for row in conn.execute("PRAGMA table_info(action_outcome_events)")
+    }
+    assert columns == {
+        "id", "objective_ref", "action_name", "outcome", "error_fingerprint",
+        "recovered_from", "sequence", "source_kind", "source_id",
+        "authority_class", "provenance_version", "created_at", "expires_at",
+    }
+
+    with pytest.raises(ValueError, match="bounded identifier"):
+        action_outcomes.record_evaluation_event(
+            conn,
+            event_id="ao-secret",
+            objective_ref="objective-1",
+            action_name="github.pr.create --body raw-output",
+            outcome="success",
+            sequence=1,
+            source_kind="dialog",
+            source_id="dialog-1",
+            authority="observed",
+            principal="fixture-runner",
+            channel="offline-evaluation",
+        )
+    with pytest.raises(RuntimeError, match="offline evaluation"):
+        action_outcomes.record_live_event()
+
+
+def test_recovery_order_and_authorized_candidate_are_bounded(fresh_mp):
+    from threadkeeper import action_outcomes
+
+    conn = fresh_mp["db"].get_db()
+    failed, recovered = _record_pair(conn, action_outcomes)
+    candidate_id = action_outcomes.derive_recovery_candidate(conn, recovered)
+
+    row = conn.execute(
+        "SELECT content, rationale, source_uuid FROM extract_candidates WHERE id=?",
+        (candidate_id,),
+    ).fetchone()
+    assert row["rationale"] == "operational_recovery"
+    assert row["source_uuid"] == f"action-outcome:{recovered}"
+    assert row["content"] == (
+        "Operational recovery: retry github.pr.create after "
+        "http.422.validation."
+    )
+    assert action_outcomes.candidate_source(conn, candidate_id) == (
+        "action_outcome_candidate", str(candidate_id)
+    )
+    exported = action_outcomes.export_events(conn)
+    assert [row["id"] for row in exported] == [failed, recovered]
+    assert set(exported[0]) == {
+        "id", "objective_ref", "action_name", "outcome", "error_fingerprint",
+        "recovered_from", "sequence", "source_kind", "source_id",
+        "authority_class", "provenance_version", "created_at", "expires_at",
+    }
+
+    with pytest.raises(ValueError, match="earlier failure"):
+        action_outcomes.record_evaluation_event(
+            conn,
+            event_id="ao-out-of-order",
+            objective_ref="objective-42",
+            action_name="github.pr.create",
+            outcome="success",
+            sequence=1,
+            recovered_from=failed,
+            source_kind="dialog",
+            source_id="dialog-42",
+            authority="observed",
+            principal="fixture-runner",
+            channel="offline-evaluation",
+        )
+    with pytest.raises(ValueError, match="same objective"):
+        action_outcomes.record_evaluation_event(
+            conn,
+            event_id="ao-cross-objective",
+            objective_ref="objective-43",
+            action_name="github.pr.create",
+            outcome="success",
+            sequence=2,
+            recovered_from=failed,
+            source_kind="dialog",
+            source_id="dialog-43",
+            authority="observed",
+            principal="fixture-runner",
+            channel="offline-evaluation",
+        )
+
+
+def test_expiry_and_source_or_objective_forgetting_remove_candidates_and_invalidate_lessons(
+    fresh_mp,
+):
+    from threadkeeper import action_outcomes
+    from threadkeeper.authority import source_is_live
+    from threadkeeper.tools.lessons import lesson_append
+
+    conn = fresh_mp["db"].get_db()
+    _failed, recovered = _record_pair(conn, action_outcomes)
+    candidate_id = action_outcomes.derive_recovery_candidate(conn, recovered)
+    conn.commit()
+
+    result = lesson_append(
+        title="Recover GitHub validation errors",
+        summary="Use the linked recovery evidence.",
+        body="Retry the normalized action only after the stable validation error.",
+        source=f"action_outcome_candidate:{candidate_id}",
+    )
+    assert result.startswith("ok slug=recover-github-validation-errors")
+    assert source_is_live(conn, ("lesson", "recover-github-validation-errors"))
+
+    deleted = action_outcomes.forget_objective(conn, "objective-42")
+    assert deleted == {"events": 2, "candidates": 1}
+    assert not source_is_live(conn, ("lesson", "recover-github-validation-errors"))
+    assert conn.execute("SELECT COUNT(*) FROM action_outcome_events").fetchone()[0] == 0
+    assert conn.execute("SELECT COUNT(*) FROM extract_candidates").fetchone()[0] == 0
+
+    old = action_outcomes.record_evaluation_event(
+        conn,
+        event_id="ao-expired",
+        objective_ref="objective-44",
+        action_name="github.pr.create",
+        outcome="failure",
+        error_fingerprint="http.429.rate-limit",
+        sequence=1,
+        source_kind="dialog",
+        source_id="dialog-44",
+        authority="observed",
+        principal="fixture-runner",
+        channel="offline-evaluation",
+        created_at=int(time.time()) - 3 * 86400,
+        retention_days=1,
+    )
+    assert old == "ao-expired"
+    assert action_outcomes.prune_expired(conn, int(time.time())) == {"events": 1, "candidates": 0}
+
+    action_outcomes.record_evaluation_event(
+        conn,
+        event_id="ao-source-forget",
+        objective_ref="objective-45",
+        action_name="github.pr.create",
+        outcome="failure",
+        error_fingerprint="http.429.rate-limit",
+        sequence=1,
+        source_kind="dialog",
+        source_id="dialog-45",
+        authority="observed",
+        principal="fixture-runner",
+        channel="offline-evaluation",
+    )
+    assert action_outcomes.forget_sources(conn, [("dialog", "dialog-45")]) == {
+        "events": 1,
+        "candidates": 0,
+    }
+
+    conn.execute(
+        "INSERT INTO dialog_messages "
+        "(uuid, source, role, content, created_at) VALUES "
+        "('dialog-46', 'pytest', 'user', 'erasable source', 100)",
+    )
+    action_outcomes.record_evaluation_event(
+        conn,
+        event_id="ao-forget-selector",
+        objective_ref="objective-46",
+        action_name="github.pr.create",
+        outcome="failure",
+        error_fingerprint="http.429.rate-limit",
+        sequence=1,
+        source_kind="dialog",
+        source_id="dialog-46",
+        authority="observed",
+        principal="fixture-runner",
+        channel="offline-evaluation",
+    )
+    conn.commit()
+    from threadkeeper.forget import forget_selector
+
+    assert "action_outcomes=1" in forget_selector(
+        "dialog-46", selector_type="uuid", conn=conn,
+    )
+    forget_selector("dialog-46", selector_type="uuid", dry_run=False, conn=conn)
+    assert conn.execute("SELECT COUNT(*) FROM action_outcome_events").fetchone()[0] == 0
+
+
+def test_offline_comparison_sets_thresholds_without_enabling_live_ingest():
+    from threadkeeper import action_outcomes
+
+    fixture = {
+        "dialogue_only": [
+            "tool arguments very-sensitive-arguments tool output unbounded-output " * 20,
+        ],
+        "events": [
+            {
+                "id": "failure", "objective_ref": "objective-1",
+                "action_name": "github.pr.create", "outcome": "failure",
+                "error_fingerprint": "http.422.validation", "sequence": 1,
+            },
+            {
+                "id": "recovery", "objective_ref": "objective-1",
+                "action_name": "github.pr.create", "outcome": "success",
+                "sequence": 2, "recovered_from": "failure",
+            },
+        ],
+        "expected_recoveries": [["failure", "recovery"]],
+        "forbidden_substrings": ["very-sensitive-arguments", "unbounded-output"],
+        "thresholds": {
+            "min_lesson_usefulness": 1.0,
+            "max_privacy_leakage": 0,
+            "max_storage_ratio": 0.5,
+            "max_false_causal_links": 0,
+        },
+    }
+    report = action_outcomes.evaluate_offline_fixture(fixture)
+
+    assert report["thresholds_pass"] is True
+    assert report["bounded_action_outcomes"]["privacy_leakage"] == 0
+    assert report["dialogue_only"]["privacy_leakage"] == 2
+    assert report["bounded_action_outcomes"]["lesson_usefulness"] == 1.0
+    assert report["live_ingestion_allowed"] is False
+    assert action_outcomes.live_ingestion_allowed(True, report) is True
+    assert json.dumps(report)

threadkeeper-0.22.0/tests/test_brief_hook_size.py

@@ -0,0 +1,69 @@
+"""The SessionStart brief stays under the CLI's hook output limit."""
+from __future__ import annotations
+
+import json
+import os
+import subprocess
+from pathlib import Path
+
+HOOK = Path(__file__).resolve().parents[1] / "scripts" / "hooks" / "tk-brief.sh"
+
+
+def _run_hook(tmp_path: Path, context_lines: int) -> str:
+    fake_python = tmp_path / "python"
+    body = "\\n".join(f"rule {i}: " + "x" * 60 for i in range(context_lines))
+    fake_python.write_text(
+        "#!/bin/sh\ncat >/dev/null\n"
+        f"printf 'STATUS\\tthread-keeper: ok\\nCONTEXT\\t{body}\\n'\n",
+        encoding="utf-8",
+    )
+    fake_python.chmod(0o700)
+    env = {
+        "PATH": os.environ.get("PATH", "/usr/bin:/bin"),
+        "HOME": str(tmp_path),
+        "THREADKEEPER_PYTHON": str(fake_python),
+    }
+    proc = subprocess.run(
+        ["bash", str(HOOK)], input="{}", capture_output=True, text=True, env=env, timeout=30,
+    )
+    assert proc.returncode == 0, proc.stderr
+    return json.loads(proc.stdout)["hookSpecificOutput"]["additionalContext"]
+
+
+def test_oversized_brief_is_cut_at_a_line_break_under_the_hook_limit(tmp_path):
+    context = _run_hook(tmp_path, context_lines=400)
+
+    assert len(context) <= 9000
+    assert context.endswith("call brief() for the rest]")
+    assert "\nrule 0: " in context
+    kept = context.splitlines()[-2]
+    assert kept.startswith("rule ") and kept.endswith("x" * 60)
+
+
+def test_brief_under_the_limit_is_passed_through_whole(tmp_path):
+    context = _run_hook(tmp_path, context_lines=20)
+
+    assert "rule 19: " in context
+    assert "trimmed" not in context
+
+
+def test_brief_without_line_breaks_is_cut_at_the_limit(tmp_path):
+    fake_python = tmp_path / "python"
+    fake_python.write_text(
+        "#!/bin/sh\ncat >/dev/null\n"
+        "printf 'STATUS\\tok\\nCONTEXT\\t%s\\n' \"$(head -c 20000 /dev/zero | tr '\\0' x)\"\n",
+        encoding="utf-8",
+    )
+    fake_python.chmod(0o700)
+    env = {
+        "PATH": os.environ.get("PATH", "/usr/bin:/bin"),
+        "HOME": str(tmp_path),
+        "THREADKEEPER_PYTHON": str(fake_python),
+    }
+    proc = subprocess.run(
+        ["bash", str(HOOK)], input="{}", capture_output=True, text=True, env=env, timeout=30,
+    )
+
+    context = json.loads(proc.stdout)["hookSpecificOutput"]["additionalContext"]
+    assert 8000 < len(context) <= 9000
+    assert context.endswith("call brief() for the rest]")

{threadkeeper-0.20.0 → threadkeeper-0.22.0}/tests/test_brief_sections.py

@@ -146,6 +146,49 @@ def test_consulted_skills_silent_without_events(mp_with_cid):
     assert "consulted_skills" not in txt
 
 
+def test_consulted_lessons_surface_recall_telemetry_and_keep_brief_sections(
+    mp_with_cid,
+):
+    """A recalled lesson keeps its read telemetry visible without dropping
+    the existing thread, evolve, or user-facing brief sections."""
+    pkg = mp_with_cid(_FAKE_CID)
+    lesson_append = _tool(pkg, "lesson_append")
+    lesson_list = _tool(pkg, "lesson_list")
+    lesson_get = _tool(pkg, "lesson_get")
+    open_thread = _tool(pkg, "open_thread")
+
+    assert lesson_append(
+        title="Recall telemetry", body="Use this lesson for brief coverage.",
+        source="shadow",
+    ).startswith("ok")
+    lesson_list(k=1)
+    assert "Use this lesson for brief coverage." in lesson_get(
+        slug="recall-telemetry"
+    )
+    open_thread(question="keep the open section")
+
+    conn = pkg["db"].get_db()
+    conn.execute(
+        "INSERT INTO evolve (suggestion, rationale, applied, status, created_at) "
+        "VALUES (?, ?, 0, 'promoted', ?)",
+        ("seeded promoted suggestion", "brief coverage", int(time.time())),
+    )
+    conn.commit()
+
+    txt = _brief_text(pkg)
+
+    assert "consulted_lessons" in txt
+    assert "recall-telemetry" in txt
+    assert "viewed×1" in txt
+    assert "used×1" in txt
+    assert "last_recalled=" in txt
+    assert "open" in txt
+    assert "keep the open section" in txt
+    assert "evolve_pending" in txt
+    assert "★ \"seeded promoted suggestion\"" in txt
+    assert "⚠️ user-facing: paraphrase plain." in txt
+
+
 def test_skill_nudge_silent_after_materialization(mp_with_cid, monkeypatch):
     monkeypatch.setenv("THREADKEEPER_SKILL_NUDGE_INTERVAL", "3")
     pkg = mp_with_cid(_FAKE_CID)