chore(governance): sync cheatsheet parity for post-holdout probes

Add EN/zh-TW cheatsheet cues for ROUTE-002 design-comparison holdout
and ROUTE-003 dispatch-meta trap; export probe tuples with anti-drift
tests. Refresh QUALITY_GATES pytest floor (180+) with collection guard.

Author: johnteee

reflective-prompt-library/PROJECT_KNOWLEDGE.md

@@ -81,6 +81,7 @@ deferred promotions are recurrence-gated — see [panel backlog](plans/multi-age
 - 2026-06-18 Adopt STORM's source-grounded perspective discovery as an optional `reflective-research` method, not a new skill → [record](plans/storm-perspective-discovery-reflection-2026-06-18.md)
 - 2026-06-20 Synthesize external-adoption case studies, then correct the evidence and promotion-gate scope after the Test Plan routing counterexample → [record](plans/external-adoption-case-studies-2026-06-20.md)
 - 2026-06-21 Evaluate Hyperplan / multi-agent adversarial planning — no change (runtime = non-goal; methodology overlaps; possible gaps not promoted) → [record](plans/external-adoption-case-studies-2026-06-20.md)
+- 2026-06-25 Post-Round 68 cheatsheet parity — ROUTE-002 design-comparison holdout phrase + ROUTE-003 dispatch-meta probe in EN/zh-TW cheatsheets → [QUALITY_GATES_SUMMARY.md](plans/QUALITY_GATES_SUMMARY.md)
 - 2026-06-25 Post-Round 68 holdout expansion — ROUTE-002 design comparison phrase; ROUTE-003 approved-spec plan-only trap + dispatch routing probe → [ROUTING_CONTRACT.md](plans/ROUTING_CONTRACT.md#r11-approved-spec-delivery)
 - 2026-06-25 Round 68 panel — full-doc drift sync, README/methodology-map anti-drift, reject holdout expansion at 100% → [record](plans/multi-agent-panel-consensus-2026-06-25.md)
 - 2026-06-25 Post-Round 68 maintenance — nine-skill wording parity (METHODOLOGY_MAP, skill-map, QUALITY_GATES), benchmark B024 dispatch (24 tasks, 9/9 workflows) → [QUALITY_GATES_SUMMARY.md](plans/QUALITY_GATES_SUMMARY.md)

reflective-prompt-library/plans/QUALITY_GATES_SUMMARY.md

@@ -314,7 +314,7 @@ ROUTE-002 measures unseen phrasing separately from ROUTE-001. Round 7 (2026-06-2
 2. **ROUTE-001/002/003 in CI** — 128 + 101 + 35 paraphrases at 100% consistency (seeded fixtures); `validate_route_fixture.py` gates minimum coverage
 3. **Governance validators** — links, lint, governance metadata, PROJECT_KNOWLEDGE, benchmark fixture, skill examples
 4. **Harness policy docs** — CONTRIBUTING, AGENTS, SKILL_INSTALLATION, maintenance playbook
-5. **Doc anti-drift** — `test_routing_contract.py`, cheatsheet parity tests, `test_readme_governance.py` (170+ pytest in CI)
+5. **Doc anti-drift** — `test_routing_contract.py`, cheatsheet parity tests, `test_readme_governance.py` (180+ pytest anti-drift suite in CI)
 
 ### Ongoing maintenance (not blockers)
 

reflective-prompt-library/plans/tests/test_cheatsheet_dispatch_meta_parity.py

@@ -0,0 +1,38 @@
+"""Anti-drift checks for dispatch-meta ROUTE-003 cues in EN/zh-TW cheatsheets."""
+
+import sys
+from pathlib import Path
+
+import pytest
+
+sys.path.insert(0, str(Path(__file__).parent))
+
+from test_validate_route_fixture import DISPATCH_META_SKILL_TRAP_PROBES  # noqa: E402
+
+SKILLS = Path(__file__).parent.parent.parent / "skills"
+EN_CHEATSHEET = SKILLS / "SKILL_TRIGGER_CHEATSHEET.md"
+ZH_CHEATSHEET = SKILLS / "SKILL_TRIGGER_CHEATSHEET.zh-TW.md"
+
+
+@pytest.fixture(scope="module")
+def en_cheatsheet_text() -> str:
+    assert EN_CHEATSHEET.is_file(), f"missing {EN_CHEATSHEET}"
+    return EN_CHEATSHEET.read_text(encoding="utf-8")
+
+
+@pytest.fixture(scope="module")
+def zh_cheatsheet_text() -> str:
+    assert ZH_CHEATSHEET.is_file(), f"missing {ZH_CHEATSHEET}"
+    return ZH_CHEATSHEET.read_text(encoding="utf-8")
+
+
+@pytest.mark.parametrize("cue", DISPATCH_META_SKILL_TRAP_PROBES)
+def test_dispatch_meta_cues_present_in_english_cheatsheet(en_cheatsheet_text: str, cue: str):
+    haystack = en_cheatsheet_text.lower()
+    assert cue.lower() in haystack, f"English cheatsheet missing dispatch-meta cue: {cue!r}"
+
+
+@pytest.mark.parametrize("cue", DISPATCH_META_SKILL_TRAP_PROBES)
+def test_dispatch_meta_cues_present_in_zh_tw_cheatsheet(zh_cheatsheet_text: str, cue: str):
+    haystack = zh_cheatsheet_text.lower()
+    assert cue.lower() in haystack, f"zh-TW cheatsheet missing dispatch-meta cue: {cue!r}"

reflective-prompt-library/plans/tests/test_quality_gates_summary.py

@@ -126,7 +126,36 @@ def test_benchmark_section_covers_nine_workflows(summary_text: str):
     assert "8 different skills" not in section
     assert "24 golden tasks" in section or "24 benchmark tasks" in section
 
+
+
+def _pytest_item_count() -> int:
+    import subprocess
+
+    result = subprocess.run(
+        ["python3", "-m", "pytest", str(Path(__file__).parent), "--collect-only", "-q"],
+        capture_output=True,
+        text=True,
+        check=True,
+    )
+    for line in reversed(result.stdout.splitlines()):
+        if " tests collected" in line:
+            return int(line.split()[0])
+    raise AssertionError("could not determine pytest collection count")
+
+
 def test_key_metrics_benchmark_task_count_matches_fixture(summary_text: str):
     metrics = _section_between(summary_text, "## Key Metrics", "### Routing Consistency Tracking")
     assert f"| Benchmark tasks | {MIN_TASK_COUNT} |" in metrics
 
+def test_phase2_pytest_floor_matches_collection(summary_text: str):
+    """Anti-drift: Phase 2 Done bullet must not under-report collected pytest count."""
+    collected = _pytest_item_count()
+    match = re.search(r"(\d+)\+ pytest anti-drift suite in CI", summary_text)
+    assert match, "missing pytest floor in Phase 2 Done bullet"
+    documented = int(match.group(1))
+    assert documented <= collected, (
+        f"QUALITY_GATES documents {documented}+ pytest but {collected} collected"
+    )
+    assert documented >= collected - 20, (
+        f"QUALITY_GATES pytest floor {documented} is stale vs {collected} collected"
+    )

reflective-prompt-library/plans/tests/test_validate_route_fixture.py

@@ -66,12 +66,16 @@ def test_route_fixture_minimums_match_validator_constants():
 )
 
 # ROUTE-002 holdout phrases that should appear in EN/zh-TW cheatsheets (anti-drift).
+DESIGN_COMPARISON_HOLDOUT_PROBE = (
+    "compare API designs on paper without touching the repository",
+)
+
 BOUNDARY_CHEATSHEET_CUES = tuple(text for text, _ in ROUND_51_BOUNDARY_PROBES) + (
     "write tickets and acceptance criteria without touching the repo",
     "check the diff for readability not production deploy",
     "審查 README 清晰度不是安全風險",
     "把規格寫出來但不要改程式",
-)
+) + DESIGN_COMPARISON_HOLDOUT_PROBE
 
 
 def test_round_51_boundary_probes():
@@ -97,6 +101,10 @@ def test_round_51_boundary_probes():
     "plan 已核准 spec without repo changes",
 )
 
+DISPATCH_META_SKILL_TRAP_PROBES = (
+    "which reflective workflow skill fits a routing-only mixed intent",
+)
+
 ROUTE_003_ADVERSARIAL_BOUNDARY_PROBES = (
     ("design a handoff workflow specification without runtime code", "reflective-spec-plan"),
     ("patch the trivial null check in code", "reflective-implement"),
@@ -222,3 +230,17 @@ def test_approved_spec_plan_not_implement_trap_covers_probes():
     missing = [p for p in IMPLEMENT_NOT_PLAN_SPEC_PLAN_PROBES if p not in fixture_phrases]
     assert not missing, f"approved_spec_plan_not_implement_trap missing probes: {missing}"
 
+def _dispatch_meta_skill_trap_phrases() -> set[str]:
+    config = load_route_eval_config(PLANS / "route-003-adversarial-eval.yaml")
+    for group in config["adversarial_sets"]:
+        if group["name"] == "dispatch_meta_skill_trap":
+            return set(group.get("phrases", []))
+    raise AssertionError("missing ROUTE-003 dispatch_meta_skill_trap group")
+
+
+def test_dispatch_meta_skill_trap_covers_probes():
+    """Anti-drift: ROUTE-003 dispatch-meta probes live in dedicated trap group."""
+    fixture_phrases = _dispatch_meta_skill_trap_phrases()
+    missing = [p for p in DISPATCH_META_SKILL_TRAP_PROBES if p not in fixture_phrases]
+    assert not missing, f"dispatch_meta_skill_trap missing probes: {missing}"
+

reflective-prompt-library/skills/SKILL_TRIGGER_CHEATSHEET.md

@@ -42,6 +42,7 @@ Trigger cues:
 - "Which skill handles session handoff in this library?"
 - "Which workflow skill should run for this library task?"
 - "Which reflective workflow skill covers handoff retro?"
+- "Which reflective workflow skill fits a routing-only mixed intent?"
 
 Do not use when:
 
@@ -86,6 +87,7 @@ Trigger cues:
 
 - "Write tickets and acceptance criteria without touching the repo."
 - "Compare two API designs on paper only no implementation."
+- "Compare API designs on paper without touching the repository."
 - "Design comparison on paper without repo changes."
 - "比較兩個 API 設計方案但不要寫 code."
 - "把規格寫出來但不要改程式."

reflective-prompt-library/skills/SKILL_TRIGGER_CHEATSHEET.zh-TW.md

@@ -42,6 +42,7 @@ Skill 契約語言：
 - 「Which skill handles session handoff in this library？」
 - 「Which workflow skill should run for this library task？」
 - 「Which reflective workflow skill covers handoff retro？」
+- 「Which reflective workflow skill fits a routing-only mixed intent？」
 
 不要用在：
 
@@ -82,6 +83,7 @@ L1 快速路徑：
 
 - 「Write tickets and acceptance criteria without touching the repo。」
 - 「Compare two API designs on paper only no implementation。」
+- 「Compare API designs on paper without touching the repository。」
 - 「比較兩個 API 設計方案但不要寫 code。」
 - 「把規格寫出來但不要改程式。」
 - 「Design a handoff workflow specification without runtime code。」