[tommy-ca]

Ref #649 (Parent Epic #645).

Summary

Remediates oversized test files flagged during the thermo-nuclear review by decomposing suites exceeding the 1,000-line ceiling:

1. Decomposed tests/test_verify_harness.py (1,437 lines):

   • tests/test_verify_harness_archive.py (157 lines): Intent archive chains, durable specs, Benny merge/push guards, and OpenSpec schema validation.
   • tests/test_verify_harness_guides.py (311 lines): Sync/adapt guide assertions, README locale split, first-session sandbox/reload, and forge-neutral PR paths.
   • tests/test_verify_harness_playbooks.py (352 lines): Autopilot fanout, host mapping requirements, bot UI invocability, Benny remap, and no-write agent capability modes.
   • tests/test_verify_harness.py (595 lines): Core harness invariant checks, manifest qualifier checks, and profile parity.

2. Extracted Shared Fixtures:

   • tests/evidence_test_helpers.py (362 lines): Centralized receipt, profile, disposable copy, and mock generators.

3. Decomposed tests/test_evidence_durability.py (1,016 lines):

   • tests/test_evidence_index_resilience.py (358 lines): Index shape validation, corruption recovery, stat-denied safety, and missing index reconstruction.
   • tests/test_evidence_durability.py (585 lines): Core receipt schema durability, profile freshness, and plane invalidation.

4. Decomposed tests/test_evidence_settlement.py (1,003 lines):

   • Reduced to 762 lines using tests.evidence_test_helpers, retaining 100% backwards-compatible re-exports.

5. Updated Check Mappings & CI:

   • Updated scripts/select_checks.py, tests/test_select_checks.py, and .github/workflows/portability-offline.yml.
   • Zero code or test files in the entire repository now exceed 1,000 lines.

[tommy-ca]

Independent Verification Report: PR #653 (Refs #649, Program #645)

Verdict: PASS

1. Commit and Branch Verification

• Target Branch: main
• Feature Branch: feat/decompose-oversized-tests-649
• Base Commit: 38654cae6ee14c3899379cc82893ed4dca3e5e5f (38654ca on main)
• Head Commit: 9b86fafb672c50a3c15572386ca4eb702156731c (9b86faf on feat/decompose-oversized-tests-649)
• Working Tree: Clean (git status reports nothing to commit, working tree clean)

---

2. Line Count and Modularity Analysis

All decomposed suites and helper modules were verified with wc -l. Every single file is strictly under 1,000 lines:

File	Line Count	Status (<1,000 lines)
tests/test_verify_harness_archive.py	157	PASS
tests/test_verify_harness_guides.py	311	PASS
tests/test_verify_harness_playbooks.py	352	PASS
tests/test_verify_harness.py	595	PASS
tests/evidence_test_helpers.py	362	PASS
tests/test_evidence_index_resilience.py	358	PASS
tests/test_evidence_durability.py	585	PASS
tests/test_evidence_settlement.py	762	PASS

Repository-Wide Scan:
A scan across all repository code and test files (.py, .sh, .ts, .js, excluding vendor/node_modules) confirmed 0 files exceed 1,000 lines. The largest code file in the entire repository is tests/test_model_role_policy.py at 995 lines.

---

3. Test Execution Receipts

1. Bun Test Suite (skills/poteto-mode/scripts):
   • Command: bun test
   • Result: 102 passed, 0 failed, 373 expect() calls across 7 files (11.77s)
2. Pytest Decomposed & Select Checks Suite:
   • Command: .venv/bin/pytest tests/test_verify_harness*.py tests/test_evidence_durability.py tests/test_evidence_index_resilience.py tests/test_evidence_settlement.py tests/test_select_checks.py
   • Result: 143 passed in 25.38s
     • tests/test_verify_harness.py: 18 passed
     • tests/test_verify_harness_archive.py: 7 passed
     • tests/test_verify_harness_guides.py: 5 passed
     • tests/test_verify_harness_playbooks.py: 10 passed
     • tests/test_evidence_durability.py: 26 passed
     • tests/test_evidence_index_resilience.py: 17 passed
     • tests/test_evidence_settlement.py: 32 passed
     • tests/test_select_checks.py: 28 passed
3. Portability Evidence Staleness:
   • Command: .venv/bin/python scripts/verify-portable.py check-staleness --host all
   • Result: ALL FRESH across grok, codex, omp, opencode, antigravity, droid
4. Portability Doctor:
   • Command: .venv/bin/python scripts/verify-portable.py doctor --host all
   • Result: 6-Harness Matrix Verdict: PASS
5. Harness Verification:
   • Command: python3 scripts/verify-harness.py
   • Result: PASS (playbooks: 22 named + opening-a-pr, principles: 24, plugin.json name: pstack)
6. Host Boundary Scanner:
   • Command: python3 scripts/scan-host-boundary.py --check
   • Result: PASS: Scanned 143 files across portable surfaces. 0 boundary leaks.
7. Project Package Consistency:
   • Command: .venv/bin/python scripts/project-package.py --check
   • Result: PASS (all manifests and models in sync; exit 0)
8. OpenSpec Specification Validation:
   • Command: openspec validate --changes --strict
   • Result: 2 passed, 0 failed (2 items)
9. Canonical Index Conformance:
   • Command: python3 scripts/canonical-index.py --check
   • Result: PASS: Canonical inventory conforms to pin.
10. Canonical Index Drift Check:
    • Command: python3 scripts/canonical-index.py --drift
    • Result: Upstream Freshness: CURRENT (0 commits ahead)

---

4. Selection Table & CI Workflow Verification

• scripts/select_checks.py: Mapped scripts/verify-harness.py to the 4 partitioned verify harness test files, and mapped scripts/evidence_durability.py and scripts/evidence_index.py to test_evidence_index_resilience.py.
• tests/test_select_checks.py: Updated selection expectations accordingly.
• .github/workflows/portability-offline.yml: Included tests/test_evidence_index_resilience.py in test matrix.

PR #653 cleanly achieves all modularity and verification criteria with zero regressions.