[CAT-UNVERIFIED tag](4) cat-mode and the prove-it skills name the tag

[EdbertChan]

Summary

cat-mode, principle-prove-it and prove-it-ship-gate now name the {{CAT-UNVERIFIED}} tag instead of bare UNVERIFIED:.

cat-mode is always loaded. Its line telling agents to write the retired marker reached every session and sent them into a block.

Each skill's test examples are updated, and test_cat_mode.py checks the skill and its verify reference.

Review Claim

Approve three corpus skills naming the tag, with their tests updated to match.

Review Lane

behavior

Review Unit

corpus-lesson

Safety Invariant

Prose and test examples only; no hook, script or install change.

Slice Rationale

Corpus skills are their own review unit and cannot share a PR with engine-runtime.

Non-goals

No hook, script or install change.

Test Plan

Test Plan

• python3 engine/skills/make-pr/scripts/preflight.py --base <this PR's base>

unit    corpus-lesson: 7 file(s)
neutral 1 file(s): tests/test_cat_mode.py
declare Review Unit: corpus-lesson
gate    python3 scripts/check_codify_has_code.py --base stack/EdbertChan/cat-unverified-tag/3-engine-prose
        ok      codify-has-code
gate    python3 scripts/check_no_dated_provenance.py --base stack/EdbertChan/cat-unverified-tag/3-engine-prose
        ok      no dated provenance
gate    python3 scripts/check_skills_three_harnesses.py
        ok      skills three-harness check
gate    python3 scripts/check_ecosystem_boundaries.py
        ok      ecosystem boundaries
gate    python3 scripts/check_skill_file_refs.py
        ok      skill file refs
gate    python3 scripts/check_skill_test_coverage.py --base stack/EdbertChan/cat-unverified-tag/3-engine-prose --head HEAD
        ok      skill test coverage
gate    python3 scripts/check_skill_trigger_mechanism.py
        ok      skill trigger mechanism
gate    python3 scripts/check_skill_trigger_policy.py
        ok	skill trigger policy
gate    python3 scripts/check_subagent_scope_contract.py
        ok	subagent scope contract
gate    python3 scripts/run_skill_scenarios.py
        ok  	subagent-fact-attributed
        ok  	stale-green-caught-without-any-admission
        ok  	stale-green-already-owned-by-the-reply
        ok  	red-then-green-is-a-fix-not-a-stale-claim
        
        ok	all 19 scenario(s) behaved as declared
ok      preflight passed

• bash scripts/run_all_tests.sh -- exit 0, 45 suites OK, 0 failed, 1677 tests

Revert Plan

Revert Plan

• Safe to revert? Yes
• Revert command: git revert <sha>
• Post-revert steps: None
• Data migration? No

🤖 Generated with Claude Code

https://claude.ai/code/session_01VKsvxJk65w6q7KnPSRYvNg

---

Note

Low Risk
Corpus prose and test examples only; no hook, install, or runtime behavior changes in this PR.

Overview
Replaces the retired bare UNVERIFIED: escape hatch with the well-formed {{CAT-UNVERIFIED: …}} tag across always-loaded cat-mode (including references/verify.md), principle-prove-it, and prove-it-ship-gate, plus their scenario test examples.

Because cat-mode loads every session, prose that still told agents to write bare UNVERIFIED: was steering them into Stop-hook blocks that only accept the structured tag (with claim and blocker). Live-path / ship claims now say to tag the claim and name the blocker instead of prefixing UNVERIFIED: live path; hedge rules note the bare marker is retired but still triggers prove-it.

Adds TestEscapeHatchVocabulary in test_cat_mode.py, which asserts cat-mode and verify.md mention CAT-UNVERIFIED and do not instruct the retired marker (via escape_hatch_vocab.instructs_retired_marker).

^{Reviewed by Cursor Bugbot for commit 529e75c. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_be1bebc5-da7f-4189-a64b-2bcc6604a4e1)

[EdbertChan]

Mergify repair stopped: lowest open stack PR #418 is based on stack/EdbertChan/cat-unverified-tag/3-engine-prose, which still belongs to open PR(s) #417 outside this stack; leaving it alone to avoid dropping dependency changes

[cursor]

Bugbot couldn't run - usage limit reached

Bugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit.

A user or team admin can review and increase usage limits in the Cursor dashboard.

(requestId: serverGenReqId_7145a1d1-c6ca-471e-805b-6e68246d26cd)

[mergify]

Tick the box to add this pull request to the merge queue (same as @mergifyio queue).

• Queue this pull request