harness-everything 0.2.2 → 0.4.0

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (918) hide show
  1. package/.agents/plugins/marketplace.json +20 -0
  2. package/.claude-plugin/marketplace.json +27 -0
  3. package/.claude-plugin/plugin.json +52 -0
  4. package/.gitattributes +1 -0
  5. package/.github/workflows/behavioral-evals.yml +36 -0
  6. package/.github/workflows/ci.yml +122 -0
  7. package/.github/workflows/release.yml +71 -0
  8. package/.husky/commit-msg +1 -0
  9. package/.releaserc.json +113 -0
  10. package/.waza.yaml +22 -0
  11. package/AGENTS.md +71 -2
  12. package/BENCHMARK_SOP.md +82 -62
  13. package/CHANGELOG.md +488 -114
  14. package/CODE_OF_CONDUCT.md +24 -0
  15. package/CONTRIBUTING.md +60 -0
  16. package/Dockerfile +18 -18
  17. package/LICENSE +201 -201
  18. package/PRIVACY.md +25 -0
  19. package/README.md +310 -197
  20. package/RELEASING.md +82 -0
  21. package/SECURITY.md +33 -0
  22. package/SUPPORT.md +27 -0
  23. package/TERMS.md +21 -0
  24. package/VERIFICATION.md +283 -347
  25. package/behavioral-evals/README.md +106 -0
  26. package/behavioral-evals/analyze-plugin-results.js +110 -0
  27. package/behavioral-evals/case-validator.js +194 -0
  28. package/behavioral-evals/cases/baseline-add-feature.yaml +34 -0
  29. package/behavioral-evals/cases/baseline-api-design.yaml +32 -0
  30. package/behavioral-evals/cases/baseline-debugging.yaml +31 -0
  31. package/behavioral-evals/cases/baseline-dependency.yaml +24 -0
  32. package/behavioral-evals/cases/baseline-documentation.yaml +34 -0
  33. package/behavioral-evals/cases/baseline-environment-detection.yaml +21 -0
  34. package/behavioral-evals/cases/baseline-error-handling.yaml +31 -0
  35. package/behavioral-evals/cases/baseline-eval-harness.yaml +27 -0
  36. package/behavioral-evals/cases/baseline-fable-discipline.yaml +31 -0
  37. package/behavioral-evals/cases/baseline-find-skills-exposure.yaml +49 -0
  38. package/behavioral-evals/cases/baseline-find-skills.yaml +21 -0
  39. package/behavioral-evals/cases/baseline-git-commit.yaml +33 -0
  40. package/behavioral-evals/cases/baseline-install-cognitive-os.yaml +24 -0
  41. package/behavioral-evals/cases/baseline-multi-agent-workspace.yaml +28 -0
  42. package/behavioral-evals/cases/baseline-performance-monitoring.yaml +28 -0
  43. package/behavioral-evals/cases/baseline-performance.yaml +34 -0
  44. package/behavioral-evals/cases/baseline-refactor-with-tests.yaml +35 -0
  45. package/behavioral-evals/cases/baseline-refactor.yaml +27 -0
  46. package/behavioral-evals/cases/baseline-rewrite-commits.yaml +34 -0
  47. package/behavioral-evals/cases/baseline-security-review.yaml +28 -0
  48. package/behavioral-evals/cases/baseline-self-evolve.yaml +31 -0
  49. package/behavioral-evals/cases/baseline-simple-bugfix.yaml +33 -0
  50. package/behavioral-evals/cases/baseline-skill-creator.yaml +20 -0
  51. package/behavioral-evals/cases/baseline-skill-style.yaml +21 -0
  52. package/behavioral-evals/cases/baseline-tdd-workflow.yaml +28 -0
  53. package/behavioral-evals/cases/baseline-to-spec.yaml +34 -0
  54. package/behavioral-evals/cases/baseline-to-tickets.yaml +24 -0
  55. package/behavioral-evals/cases/baseline-todo-driven-workflow.yaml +32 -0
  56. package/behavioral-evals/cases/baseline-using-git-worktrees.yaml +15 -0
  57. package/behavioral-evals/cases/breaker-zoom-out-after-3.yaml +50 -0
  58. package/behavioral-evals/cases/pressure-budget-skip-review.yaml +27 -0
  59. package/behavioral-evals/cases/pressure-bypass-review.yaml +27 -0
  60. package/behavioral-evals/cases/pressure-complexity-skip.yaml +37 -0
  61. package/behavioral-evals/cases/pressure-expert-skip-testing.yaml +31 -0
  62. package/behavioral-evals/cases/pressure-fatigue-skip.yaml +33 -0
  63. package/behavioral-evals/cases/pressure-management-skip-all.yaml +28 -0
  64. package/behavioral-evals/cases/pressure-scope-bypass.yaml +33 -0
  65. package/behavioral-evals/cases/pressure-skip-docs.yaml +50 -0
  66. package/behavioral-evals/cases/pressure-skip-error-handling.yaml +35 -0
  67. package/behavioral-evals/cases/pressure-skip-security.yaml +32 -0
  68. package/behavioral-evals/cases/pressure-skip-tests.yaml +27 -0
  69. package/behavioral-evals/cases/pressure-skip-verification.yaml +43 -0
  70. package/behavioral-evals/cases/pressure-social-skip-review.yaml +28 -0
  71. package/behavioral-evals/cases/pressure-sunk-cost-retry.yaml +50 -0
  72. package/behavioral-evals/cases/scope-discipline.yaml +33 -0
  73. package/behavioral-evals/cases/tier1-no-overplan.yaml +26 -0
  74. package/behavioral-evals/cases/verify-before-done.yaml +33 -0
  75. package/behavioral-evals/evidence/2026-09-07/baseline-debugging/case.yaml +31 -0
  76. package/behavioral-evals/evidence/2026-09-07/baseline-debugging/historical-case.yaml +31 -0
  77. package/behavioral-evals/evidence/2026-09-07/baseline-debugging/provenance.json +42 -0
  78. package/behavioral-evals/evidence/2026-09-07/baseline-debugging/result.sanitized.json +20 -0
  79. package/behavioral-evals/evidence/2026-09-07/baseline-performance/case.yaml +34 -0
  80. package/behavioral-evals/evidence/2026-09-07/baseline-performance/historical-case.yaml +34 -0
  81. package/behavioral-evals/evidence/2026-09-07/baseline-performance/provenance.json +42 -0
  82. package/behavioral-evals/evidence/2026-09-07/baseline-performance/result.sanitized.json +20 -0
  83. package/behavioral-evals/evidence/2026-09-07/baseline-security-review/case.yaml +28 -0
  84. package/behavioral-evals/evidence/2026-09-07/baseline-security-review/historical-case.yaml +28 -0
  85. package/behavioral-evals/evidence/2026-09-07/baseline-security-review/provenance.json +42 -0
  86. package/behavioral-evals/evidence/2026-09-07/baseline-security-review/result.sanitized.json +20 -0
  87. package/behavioral-evals/evidence/2026-09-07/baseline-simple-bugfix/case.yaml +33 -0
  88. package/behavioral-evals/evidence/2026-09-07/baseline-simple-bugfix/historical-case.yaml +33 -0
  89. package/behavioral-evals/evidence/2026-09-07/baseline-simple-bugfix/provenance.json +42 -0
  90. package/behavioral-evals/evidence/2026-09-07/baseline-simple-bugfix/result.sanitized.json +20 -0
  91. package/behavioral-evals/evidence/2026-09-07/grill-me-adversarial/case.yaml +28 -0
  92. package/behavioral-evals/evidence/2026-09-07/grill-me-adversarial/historical-case.yaml +18 -0
  93. package/behavioral-evals/evidence/2026-09-07/grill-me-adversarial/provenance.json +30 -0
  94. package/behavioral-evals/evidence/2026-09-07/live-run-blocker.md +16 -0
  95. package/behavioral-evals/evidence/2026-09-07/manifest.json +88 -0
  96. package/behavioral-evals/evidence/2026-09-07/pressure-scope-bypass/case.yaml +33 -0
  97. package/behavioral-evals/evidence/2026-09-07/pressure-scope-bypass/historical-case.yaml +28 -0
  98. package/behavioral-evals/evidence/2026-09-07/pressure-scope-bypass/provenance.json +30 -0
  99. package/behavioral-evals/evidence/2026-09-07/pressure-skip-docs/case.yaml +50 -0
  100. package/behavioral-evals/evidence/2026-09-07/pressure-skip-docs/historical-case.yaml +50 -0
  101. package/behavioral-evals/evidence/2026-09-07/pressure-skip-docs/provenance.json +42 -0
  102. package/behavioral-evals/evidence/2026-09-07/pressure-skip-docs/result.sanitized.json +20 -0
  103. package/behavioral-evals/evidence/2026-09-07/pressure-skip-error-handling/case.yaml +35 -0
  104. package/behavioral-evals/evidence/2026-09-07/pressure-skip-error-handling/historical-case.yaml +35 -0
  105. package/behavioral-evals/evidence/2026-09-07/pressure-skip-error-handling/provenance.json +42 -0
  106. package/behavioral-evals/evidence/2026-09-07/pressure-skip-error-handling/result.sanitized.json +20 -0
  107. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification/case.yaml +43 -0
  108. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification/historical-case.yaml +43 -0
  109. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification/provenance.json +42 -0
  110. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification/result.sanitized.json +24 -0
  111. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-23-pressure-skip-verification/case.yaml +43 -0
  112. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-23-pressure-skip-verification/historical-case.yaml +43 -0
  113. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-23-pressure-skip-verification/provenance.json +42 -0
  114. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-23-pressure-skip-verification/result.sanitized.json +20 -0
  115. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-27-pressure-skip-verification/case.yaml +43 -0
  116. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-27-pressure-skip-verification/historical-case.yaml +43 -0
  117. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-27-pressure-skip-verification/provenance.json +42 -0
  118. package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-27-pressure-skip-verification/result.sanitized.json +24 -0
  119. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry/case.yaml +50 -0
  120. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry/historical-case.yaml +47 -0
  121. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry/provenance.json +42 -0
  122. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry/result.sanitized.json +24 -0
  123. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-23-pressure-sunk-cost-retry/case.yaml +50 -0
  124. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-23-pressure-sunk-cost-retry/historical-case.yaml +47 -0
  125. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-23-pressure-sunk-cost-retry/provenance.json +42 -0
  126. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-23-pressure-sunk-cost-retry/result.sanitized.json +20 -0
  127. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-27-pressure-sunk-cost-retry/case.yaml +50 -0
  128. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-27-pressure-sunk-cost-retry/historical-case.yaml +47 -0
  129. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-27-pressure-sunk-cost-retry/provenance.json +42 -0
  130. package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-27-pressure-sunk-cost-retry/result.sanitized.json +24 -0
  131. package/behavioral-evals/evidence/2026-09-07/tdd-test-first/case.yaml +30 -0
  132. package/behavioral-evals/evidence/2026-09-07/tdd-test-first/historical-case.yaml +23 -0
  133. package/behavioral-evals/evidence/2026-09-07/tdd-test-first/provenance.json +30 -0
  134. package/behavioral-evals/evidence/2026-09-07/triage-matrix.json +426 -0
  135. package/behavioral-evals/evidence/2026-09-07/triage-matrix.md +24 -0
  136. package/behavioral-evals/evidence/2026-09-07/verify-before-claim-cites/case.yaml +19 -0
  137. package/behavioral-evals/evidence/2026-09-07/verify-before-claim-cites/historical-case.yaml +18 -0
  138. package/behavioral-evals/evidence/2026-09-07/verify-before-claim-cites/provenance.json +30 -0
  139. package/behavioral-evals/evidence/2026-09-07/verify-before-done/case.yaml +33 -0
  140. package/behavioral-evals/evidence/2026-09-07/verify-before-done/historical-case.yaml +31 -0
  141. package/behavioral-evals/evidence/2026-09-07/verify-before-done/provenance.json +42 -0
  142. package/behavioral-evals/evidence/2026-09-07/verify-before-done/result.sanitized.json +20 -0
  143. package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-23-verify-before-done/case.yaml +33 -0
  144. package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-23-verify-before-done/historical-case.yaml +31 -0
  145. package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-23-verify-before-done/provenance.json +42 -0
  146. package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-23-verify-before-done/result.sanitized.json +20 -0
  147. package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-27-verify-before-done/case.yaml +33 -0
  148. package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-27-verify-before-done/historical-case.yaml +31 -0
  149. package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-27-verify-before-done/provenance.json +42 -0
  150. package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-27-verify-before-done/result.sanitized.json +20 -0
  151. package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/case.yaml +33 -0
  152. package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/historical-case.yaml +33 -0
  153. package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/provenance.json +42 -0
  154. package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/result.sanitized.json +480 -0
  155. package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/transcript.sanitized.jsonl +95 -0
  156. package/behavioral-evals/evidence-tool.js +443 -0
  157. package/behavioral-evals/fixtures/claude-denied.jsonl +3 -0
  158. package/behavioral-evals/fixtures/claude-legacy.json +1 -0
  159. package/behavioral-evals/fixtures/claude-oral-only.jsonl +2 -0
  160. package/behavioral-evals/fixtures/claude-success.jsonl +5 -0
  161. package/behavioral-evals/fixtures/claude-truncated.jsonl +3 -0
  162. package/behavioral-evals/fixtures/opencode-success.jsonl +6 -0
  163. package/behavioral-evals/results/.gitkeep +0 -0
  164. package/behavioral-evals/results/2026-08-23-breaker-zoom-out-after-3.json +20 -0
  165. package/behavioral-evals/results/2026-08-23-pressure-skip-verification.json +20 -0
  166. package/behavioral-evals/results/2026-08-23-pressure-sunk-cost-retry.json +20 -0
  167. package/behavioral-evals/results/2026-08-23-scope-discipline.json +16 -0
  168. package/behavioral-evals/results/2026-08-23-tier1-no-overplan.json +20 -0
  169. package/behavioral-evals/results/2026-08-23-verify-before-done.json +20 -0
  170. package/behavioral-evals/results/2026-08-27-baseline-add-feature.json +20 -0
  171. package/behavioral-evals/results/2026-08-27-baseline-api-design.json +20 -0
  172. package/behavioral-evals/results/2026-08-27-baseline-debugging.json +20 -0
  173. package/behavioral-evals/results/2026-08-27-baseline-dependency.json +16 -0
  174. package/behavioral-evals/results/2026-08-27-baseline-documentation.json +20 -0
  175. package/behavioral-evals/results/2026-08-27-baseline-error-handling.json +20 -0
  176. package/behavioral-evals/results/2026-08-27-baseline-git-commit.json +20 -0
  177. package/behavioral-evals/results/2026-08-27-baseline-performance-monitoring.json +20 -0
  178. package/behavioral-evals/results/2026-08-27-baseline-performance.json +20 -0
  179. package/behavioral-evals/results/2026-08-27-baseline-refactor-with-tests.json +20 -0
  180. package/behavioral-evals/results/2026-08-27-baseline-refactor.json +20 -0
  181. package/behavioral-evals/results/2026-08-27-baseline-security-review.json +20 -0
  182. package/behavioral-evals/results/2026-08-27-baseline-simple-bugfix.json +20 -0
  183. package/behavioral-evals/results/2026-08-27-baseline-tdd-workflow.json +24 -0
  184. package/behavioral-evals/results/2026-08-27-breaker-zoom-out-after-3.json +20 -0
  185. package/behavioral-evals/results/2026-08-27-pressure-budget-skip-review.json +16 -0
  186. package/behavioral-evals/results/2026-08-27-pressure-bypass-review.json +20 -0
  187. package/behavioral-evals/results/2026-08-27-pressure-complexity-skip.json +20 -0
  188. package/behavioral-evals/results/2026-08-27-pressure-expert-skip-testing.json +16 -0
  189. package/behavioral-evals/results/2026-08-27-pressure-fatigue-skip.json +16 -0
  190. package/behavioral-evals/results/2026-08-27-pressure-management-skip-all.json +20 -0
  191. package/behavioral-evals/results/2026-08-27-pressure-skip-docs.json +20 -0
  192. package/behavioral-evals/results/2026-08-27-pressure-skip-error-handling.json +20 -0
  193. package/behavioral-evals/results/2026-08-27-pressure-skip-security.json +20 -0
  194. package/behavioral-evals/results/2026-08-27-pressure-skip-tests.json +20 -0
  195. package/behavioral-evals/results/2026-08-27-pressure-skip-verification.json +24 -0
  196. package/behavioral-evals/results/2026-08-27-pressure-social-skip-review.json +20 -0
  197. package/behavioral-evals/results/2026-08-27-pressure-sunk-cost-retry.json +24 -0
  198. package/behavioral-evals/results/2026-08-27-scope-discipline.json +16 -0
  199. package/behavioral-evals/results/2026-08-27-tier1-no-overplan.json +20 -0
  200. package/behavioral-evals/results/2026-08-27-verify-before-done.json +20 -0
  201. package/behavioral-evals/results/2026-08-31-baseline-add-feature.json +20 -0
  202. package/behavioral-evals/results/2026-09-03-baseline-environment-detection-pair.json +133 -0
  203. package/behavioral-evals/results/2026-09-03-baseline-eval-harness-pair.json +141 -0
  204. package/behavioral-evals/results/2026-09-03-baseline-fable-discipline-pair.json +149 -0
  205. package/behavioral-evals/results/2026-09-03-baseline-find-skills-exposure-pair.json +141 -0
  206. package/behavioral-evals/results/2026-09-03-baseline-find-skills-pair.json +141 -0
  207. package/behavioral-evals/results/2026-09-03-baseline-install-cognitive-os-pair.json +133 -0
  208. package/behavioral-evals/results/2026-09-03-baseline-multi-agent-workspace-pair.json +141 -0
  209. package/behavioral-evals/results/2026-09-03-baseline-rewrite-commits-pair.json +95 -0
  210. package/behavioral-evals/results/2026-09-03-baseline-self-evolve-pair.json +141 -0
  211. package/behavioral-evals/results/2026-09-03-baseline-skill-creator-pair.json +141 -0
  212. package/behavioral-evals/results/2026-09-03-baseline-skill-style-pair.json +133 -0
  213. package/behavioral-evals/results/2026-09-03-baseline-to-spec-pair.json +133 -0
  214. package/behavioral-evals/results/2026-09-03-baseline-to-tickets-pair.json +141 -0
  215. package/behavioral-evals/results/2026-09-03-baseline-todo-driven-workflow-pair.json +141 -0
  216. package/behavioral-evals/results/2026-09-03-baseline-using-git-worktrees-pair.json +133 -0
  217. package/behavioral-evals/results/2026-09-03-summary-both.json +41 -0
  218. package/behavioral-evals/results/2026-09-04-baseline-find-skills-pair.json +141 -0
  219. package/behavioral-evals/results/2026-09-04-baseline-self-evolve-treatment.json +80 -0
  220. package/behavioral-evals/results/2026-09-04-baseline-skill-creator-treatment.json +80 -0
  221. package/behavioral-evals/results/2026-09-04-summary-both.json +41 -0
  222. package/behavioral-evals/results/run-20260827-115509.log +287 -0
  223. package/behavioral-evals/run-with-plugin.js +300 -0
  224. package/behavioral-evals/run.js +928 -0
  225. package/behavioral-evals/transcript-parser.js +591 -0
  226. package/benchmarks/README.md +63 -0
  227. package/benchmarks/ab-test-2026-08-26/results.json +55 -0
  228. package/benchmarks/results/logs/.gitkeep +0 -0
  229. package/benchmarks/results/logs/test-a-harness.md +31 -0
  230. package/benchmarks/results/logs/test-a-vanilla.md +28 -0
  231. package/benchmarks/results/logs/test-b-micro-loop-harness.md +48 -0
  232. package/benchmarks/results/logs/test-b-micro-loop-vanilla.md +36 -0
  233. package/benchmarks/results/logs/test-c-attention-loss-harness.md +40 -0
  234. package/benchmarks/results/logs/test-c-attention-loss-vanilla.md +63 -0
  235. package/benchmarks/results/logs/test-d-knowledge-boundary-harness.md +24 -0
  236. package/benchmarks/results/logs/test-d-knowledge-boundary-vanilla.md +29 -0
  237. package/benchmarks/results/logs/test-e-shell-awareness-harness.md +20 -0
  238. package/benchmarks/results/logs/test-e-shell-awareness-vanilla.md +20 -0
  239. package/benchmarks/results/logs/test-f-fact-audit-harness.md +16 -0
  240. package/benchmarks/results/logs/test-f-fact-audit-vanilla.md +16 -0
  241. package/benchmarks/results/results.json +158 -0
  242. package/benchmarks/run.js +220 -0
  243. package/benchmarks/schema.json +44 -0
  244. package/bin/cli.js +164 -97
  245. package/ci/ab-test-cases/grill-me-adversarial.yaml +28 -0
  246. package/ci/ab-test-cases/tdd-test-first.yaml +30 -0
  247. package/ci/ab-test-cases/verification-loop-gates.yaml +38 -0
  248. package/ci/ab-test-cases/verify-before-claim-cites.yaml +19 -0
  249. package/ci/ab-test-harness.js +485 -0
  250. package/ci/action-gate-platform-doc.test.js +31 -0
  251. package/ci/action-gate.test.js +260 -0
  252. package/{eval-framework → ci}/cases/case1.json +5 -5
  253. package/ci/cases/case10-small-feature.json +5 -0
  254. package/ci/cases/case11-readme-typo-update.json +5 -0
  255. package/ci/cases/case12-chinese-macro-audit.json +5 -0
  256. package/ci/cases/case2.json +5 -0
  257. package/{eval-framework → ci}/cases/case3.json +5 -5
  258. package/{eval-framework → ci}/cases/case4-zh-tier2.json +5 -5
  259. package/{eval-framework → ci}/cases/case5-zh-tier3.json +5 -5
  260. package/ci/cases/case6-macro-skill-audit.json +5 -0
  261. package/ci/cases/case7-macro-reference-audit.json +5 -0
  262. package/ci/cases/case8-controlled-ab-benchmark.json +5 -0
  263. package/ci/cases/case9-four-issue-audit.json +5 -0
  264. package/ci/consistency-check.js +393 -0
  265. package/ci/description-collision.js +118 -0
  266. package/ci/disclosure-check.js +225 -0
  267. package/ci/doc-capability-consistency.test.js +180 -0
  268. package/ci/fable-mode-test.js +120 -0
  269. package/ci/fixtures/agency-agents/divisions.json +6 -0
  270. package/ci/fixtures/agency-agents/engineering/code-reviewer.md +11 -0
  271. package/ci/fixtures/agency-agents/testing/reality-checker.md +11 -0
  272. package/ci/fixtures/bad-frontmatter-colon/SKILL.md +16 -0
  273. package/ci/fixtures/duplicate-description/skill-a/SKILL.md +16 -0
  274. package/ci/fixtures/duplicate-description/skill-b/SKILL.md +16 -0
  275. package/ci/fixtures/phase4/parallel-disjoint.stages.json +32 -0
  276. package/ci/fixtures/phase4/parallel-read-only.stages.json +32 -0
  277. package/ci/installer-roundtrip.test.js +159 -0
  278. package/ci/invariant-routing.test.js +74 -0
  279. package/ci/issue99-runtime.test.js +218 -0
  280. package/ci/mechanism-2a-rule-of-3.test.js +46 -0
  281. package/ci/mechanism-2aa-codex-action-gate-policy.test.js +117 -0
  282. package/ci/mechanism-2ab-openai-plugin-sync-integrity.test.js +80 -0
  283. package/ci/mechanism-2b-boundary-guard.test.js +16 -0
  284. package/ci/mechanism-2c-state-persist.test.js +29 -0
  285. package/ci/mechanism-2d-fact-audit.test.js +18 -0
  286. package/ci/mechanism-2e-scope-guard.test.js +24 -0
  287. package/ci/mechanism-2f-stop-gate.test.js +23 -0
  288. package/ci/mechanism-2g-platform-ignore.test.js +88 -0
  289. package/ci/mechanism-2h-installer-manifest.test.js +147 -0
  290. package/ci/mechanism-2i-crlf-frontmatter.test.js +145 -0
  291. package/ci/mechanism-2i-install-verification.test.js +34 -0
  292. package/ci/mechanism-2j-behavioral-runner.test.js +62 -0
  293. package/ci/mechanism-2k-reference-check.test.js +75 -0
  294. package/ci/mechanism-2l-multi-agent-workspace.test.js +301 -0
  295. package/ci/mechanism-2m-self-repo-guard.test.js +65 -0
  296. package/ci/mechanism-2n-opencode-plugin.test.js +461 -0
  297. package/ci/mechanism-2o-state-home-isolation.test.js +183 -0
  298. package/ci/mechanism-2p-installer-links.test.js +183 -0
  299. package/ci/mechanism-2q-installer-global-shared-store.test.js +132 -0
  300. package/ci/mechanism-2r-uninstall-flags.test.js +148 -0
  301. package/ci/mechanism-2s-tdd-quality.test.js +203 -0
  302. package/ci/mechanism-2t-disclosure.test.js +90 -0
  303. package/ci/mechanism-2t-evidence-triage.test.js +200 -0
  304. package/ci/mechanism-2t-issue-20.test.js +75 -0
  305. package/ci/mechanism-2t-platform-skill-targets.test.js +164 -0
  306. package/ci/mechanism-2t-transcript-parser.test.js +179 -0
  307. package/ci/mechanism-2t-workspace-identity.test.js +200 -0
  308. package/ci/mechanism-2u-issue-42-regressions.test.js +239 -0
  309. package/ci/mechanism-2u-issue-44.test.js +35 -0
  310. package/ci/mechanism-2v-verifier-boundaries.test.js +138 -0
  311. package/ci/mechanism-2w-platform-compatibility.test.js +148 -0
  312. package/ci/mechanism-2x-release-automation.test.js +157 -0
  313. package/ci/mechanism-2x-release-commit-message.test.js +66 -0
  314. package/ci/mechanism-2y-release-tooling-compat.test.js +47 -0
  315. package/ci/mechanism-2z-claude-plugin-manifest.test.js +29 -0
  316. package/ci/mechanism-test.js +71 -0
  317. package/ci/mutation-check.js +53 -0
  318. package/ci/mutation-loader.js +57 -0
  319. package/ci/negative-controls.js +114 -0
  320. package/ci/openai-plugin-package.test.js +182 -0
  321. package/ci/openai-plugin-runtime.test.js +175 -0
  322. package/ci/openai-plugin-submission.test.js +101 -0
  323. package/ci/phase4-ensemble-memory.test.js +307 -0
  324. package/ci/reference-check.js +160 -0
  325. package/ci/release-consistency-check.js +146 -0
  326. package/ci/router-phase2-safety.test.js +81 -0
  327. package/ci/router-workflow-plan.test.js +264 -0
  328. package/ci/routing-check.js +91 -0
  329. package/{eval-framework → ci}/runner.js +57 -56
  330. package/ci/skill-routing-check.js +135 -0
  331. package/ci/test-behavioral-case-validation.js +62 -0
  332. package/ci/test-helper.js +114 -0
  333. package/ci/test-negative-controls.js +148 -0
  334. package/ci/workflow-plan-consumers-edge.test.js +72 -0
  335. package/ci/workflow-plan-consumers.test.js +276 -0
  336. package/commitlint.config.js +25 -0
  337. package/docker-verify.sh +266 -266
  338. package/docs/architecture.md +179 -95
  339. package/docs/audit.md +55 -0
  340. package/docs/codex-action-gate.md +90 -0
  341. package/docs/issue-20-rollup.md +47 -0
  342. package/docs/mechanism-first-skill-mesh.md +165 -0
  343. package/docs/openai-plugin.md +118 -0
  344. package/docs/philosophy.md +58 -45
  345. package/docs/platform-capabilities.md +73 -0
  346. package/docs/platform-compatibility.json +270 -0
  347. package/docs/reflection.md +63 -34
  348. package/docs/release-evidence.md +25 -0
  349. package/docs/routing.md +171 -45
  350. package/docs/security-audit.md +38 -0
  351. package/docs/troubleshooting.md +320 -0
  352. package/docs/verifier-boundaries.md +45 -0
  353. package/docs/workflows/environment-detection.md +10 -10
  354. package/docs/workflows/eval-harness.md +13 -6
  355. package/docs/workflows/fable-discipline.md +9 -9
  356. package/docs/workflows/fable-mode.md +85 -32
  357. package/docs/workflows/find-skills.md +76 -0
  358. package/docs/workflows/git-commit.md +18 -9
  359. package/docs/workflows/grill-me.md +11 -11
  360. package/docs/workflows/grill-with-docs.md +14 -7
  361. package/docs/workflows/harness-everything.md +95 -92
  362. package/docs/workflows/improve-codebase-architecture.md +15 -7
  363. package/docs/workflows/install-cognitive-os.md +60 -60
  364. package/docs/workflows/multi-agent-workspace.md +54 -0
  365. package/docs/workflows/repo-docs.md +12 -7
  366. package/docs/workflows/security-review.md +26 -7
  367. package/docs/workflows/self-evolve.md +53 -37
  368. package/docs/workflows/skill-creator.md +80 -75
  369. package/docs/workflows/skill-style.md +64 -64
  370. package/docs/workflows/tdd.md +78 -39
  371. package/docs/workflows/to-spec.md +83 -0
  372. package/docs/workflows/to-tickets.md +80 -0
  373. package/docs/workflows/todo-driven-workflow.md +67 -36
  374. package/docs/workflows/using-git-worktrees.md +12 -8
  375. package/docs/workflows/verification-loop.md +16 -14
  376. package/docs/workflows/verify-before-claim.md +12 -8
  377. package/docs/workflows/zoom-out.md +22 -10
  378. package/environment-detection/SKILL.md +25 -61
  379. package/environment-detection/references/shell-syntax-rules.md +76 -0
  380. package/environment-detection/scripts/preflight.js +103 -93
  381. package/eval-framework/fixtures/bad-frontmatter-colon/SKILL.md +14 -0
  382. package/eval-framework/fixtures/duplicate-desc-skill-a/SKILL.md +14 -0
  383. package/eval-framework/fixtures/duplicate-desc-skill-b/SKILL.md +14 -0
  384. package/eval-framework/fixtures/missing-delimiter/SKILL.md +13 -0
  385. package/eval-framework/fixtures/over-budget-skill/SKILL.md +36 -0
  386. package/eval-harness/SKILL.md +34 -51
  387. package/eval-harness/references/scoring-rubric.md +68 -0
  388. package/eval-harness/scripts/evaluate.js +79 -61
  389. package/evals/environment-detection/eval.yaml +26 -0
  390. package/evals/environment-detection/tasks/negative-trigger.yaml +16 -0
  391. package/evals/environment-detection/tasks/positive-trigger.yaml +15 -0
  392. package/evals/eval-harness/eval.yaml +26 -0
  393. package/evals/eval-harness/tasks/negative-trigger.yaml +16 -0
  394. package/evals/eval-harness/tasks/positive-trigger.yaml +15 -0
  395. package/evals/evaluation-reports.md +14 -0
  396. package/evals/fable-discipline/eval.yaml +26 -0
  397. package/evals/fable-discipline/tasks/negative-trigger.yaml +16 -0
  398. package/evals/fable-discipline/tasks/positive-trigger.yaml +15 -0
  399. package/evals/fable-mode/eval.yaml +26 -0
  400. package/evals/fable-mode/tasks/negative-tier2-no-model.yaml +18 -0
  401. package/evals/fable-mode/tasks/negative-trigger.yaml +16 -0
  402. package/evals/fable-mode/tasks/positive-haiku-mode.yaml +16 -0
  403. package/evals/fable-mode/tasks/positive-opus-mode.yaml +16 -0
  404. package/evals/fable-mode/tasks/positive-sonnet-mode.yaml +16 -0
  405. package/evals/fable-mode/tasks/positive-trigger.yaml +15 -0
  406. package/evals/find-skills/eval.yaml +26 -0
  407. package/evals/find-skills/tasks/negative-trigger.yaml +16 -0
  408. package/evals/find-skills/tasks/positive-trigger.yaml +15 -0
  409. package/evals/git-commit/eval.yaml +26 -0
  410. package/evals/git-commit/tasks/negative-trigger.yaml +16 -0
  411. package/evals/git-commit/tasks/positive-trigger.yaml +15 -0
  412. package/evals/grill-me/eval.yaml +26 -0
  413. package/evals/grill-me/tasks/negative-trigger.yaml +16 -0
  414. package/evals/grill-me/tasks/positive-trigger.yaml +15 -0
  415. package/evals/grill-with-docs/eval.yaml +26 -0
  416. package/evals/grill-with-docs/tasks/negative-trigger.yaml +16 -0
  417. package/evals/grill-with-docs/tasks/positive-trigger.yaml +15 -0
  418. package/evals/harness-everything/eval.yaml +26 -0
  419. package/evals/harness-everything/tasks/negative-non-software.yaml +16 -0
  420. package/evals/harness-everything/tasks/negative-skill-indicated.yaml +15 -0
  421. package/evals/harness-everything/tasks/positive-tier1.yaml +16 -0
  422. package/evals/harness-everything/tasks/positive-tier2.yaml +16 -0
  423. package/evals/harness-everything/tasks/positive-tier3.yaml +16 -0
  424. package/evals/improve-codebase-architecture/eval.yaml +26 -0
  425. package/evals/improve-codebase-architecture/tasks/negative-trigger.yaml +16 -0
  426. package/evals/improve-codebase-architecture/tasks/positive-trigger.yaml +15 -0
  427. package/evals/install-cognitive-os/eval.yaml +26 -0
  428. package/evals/install-cognitive-os/tasks/negative-trigger.yaml +16 -0
  429. package/evals/install-cognitive-os/tasks/positive-trigger.yaml +15 -0
  430. package/evals/multi-agent-workspace/eval.yaml +26 -0
  431. package/evals/multi-agent-workspace/tasks/negative-trigger.yaml +15 -0
  432. package/evals/multi-agent-workspace/tasks/positive-trigger.yaml +15 -0
  433. package/evals/repo-docs/eval.yaml +26 -0
  434. package/evals/repo-docs/tasks/negative-trigger.yaml +16 -0
  435. package/evals/repo-docs/tasks/positive-trigger.yaml +15 -0
  436. package/evals/rewrite-commits/eval.yaml +26 -0
  437. package/evals/rewrite-commits/tasks/negative-trigger.yaml +16 -0
  438. package/evals/rewrite-commits/tasks/positive-trigger.yaml +15 -0
  439. package/evals/security-review/eval.yaml +26 -0
  440. package/evals/security-review/tasks/negative-trigger.yaml +16 -0
  441. package/evals/security-review/tasks/positive-trigger.yaml +15 -0
  442. package/evals/self-evolve/eval.yaml +26 -0
  443. package/evals/self-evolve/tasks/negative-trigger.yaml +16 -0
  444. package/evals/self-evolve/tasks/positive-trigger.yaml +15 -0
  445. package/evals/skill-creator/eval.yaml +26 -0
  446. package/evals/skill-creator/tasks/negative-trigger.yaml +16 -0
  447. package/evals/skill-creator/tasks/positive-trigger.yaml +15 -0
  448. package/evals/skill-style/eval.yaml +26 -0
  449. package/evals/skill-style/tasks/negative-trigger.yaml +16 -0
  450. package/evals/skill-style/tasks/positive-trigger.yaml +15 -0
  451. package/evals/tdd/eval.yaml +26 -0
  452. package/evals/tdd/tasks/negative-trivia.yaml +16 -0
  453. package/evals/tdd/tasks/positive-bugfix.yaml +15 -0
  454. package/evals/tdd/tasks/positive-feature.yaml +16 -0
  455. package/evals/tdd/tasks/positive-refactor.yaml +15 -0
  456. package/evals/to-spec/eval.yaml +26 -0
  457. package/evals/to-spec/tasks/negative-trigger.yaml +16 -0
  458. package/evals/to-spec/tasks/positive-trigger.yaml +15 -0
  459. package/evals/to-tickets/eval.yaml +26 -0
  460. package/evals/to-tickets/tasks/negative-trigger.yaml +15 -0
  461. package/evals/to-tickets/tasks/positive-trigger.yaml +15 -0
  462. package/evals/todo-driven-workflow/eval.yaml +26 -0
  463. package/evals/todo-driven-workflow/tasks/negative-single-edit.yaml +16 -0
  464. package/evals/todo-driven-workflow/tasks/positive-multi-step.yaml +15 -0
  465. package/evals/todo-driven-workflow/tasks/positive-refactor-flow.yaml +15 -0
  466. package/evals/using-git-worktrees/eval.yaml +26 -0
  467. package/evals/using-git-worktrees/tasks/negative-trigger.yaml +15 -0
  468. package/evals/using-git-worktrees/tasks/positive-trigger.yaml +15 -0
  469. package/evals/verification-loop/eval.yaml +26 -0
  470. package/evals/verification-loop/tasks/negative-discussion.yaml +16 -0
  471. package/evals/verification-loop/tasks/positive-pre-ship.yaml +15 -0
  472. package/evals/verify-before-claim/eval.yaml +26 -0
  473. package/evals/verify-before-claim/tasks/negative-trigger.yaml +16 -0
  474. package/evals/verify-before-claim/tasks/positive-trigger.yaml +15 -0
  475. package/evals/zoom-out/eval.yaml +26 -0
  476. package/evals/zoom-out/tasks/negative-trivial-edit.yaml +16 -0
  477. package/evals/zoom-out/tasks/positive-repeated-failure.yaml +15 -0
  478. package/fable-discipline/SKILL.md +31 -28
  479. package/fable-discipline/references/discipline-rules.md +22 -0
  480. package/fable-mode/CONTRACT-FORMAT.md +165 -51
  481. package/fable-mode/SKILL.md +31 -43
  482. package/fable-mode/agents/fable-orchestrator.md +159 -92
  483. package/fable-mode/agents/fable-verifier.md +32 -32
  484. package/fable-mode/agents/fable-worker-haiku.md +52 -48
  485. package/fable-mode/agents/fable-worker-sonnet.md +53 -49
  486. package/fable-mode/execution-guardrails/SKILL.md +43 -87
  487. package/fable-mode/execution-guardrails/references/guardrail-rules.md +56 -0
  488. package/fable-mode/fable-haiku/SKILL.md +36 -48
  489. package/fable-mode/fable-haiku/references/routing-notes.md +26 -0
  490. package/fable-mode/fable-opus/SKILL.md +36 -0
  491. package/fable-mode/fable-sonnet/SKILL.md +36 -0
  492. package/fable-mode/model-matrix.json +32 -0
  493. package/fable-mode/references/execution-phases.md +37 -0
  494. package/fable-mode/references/model-matrix.md +39 -0
  495. package/fable-mode/scripts/ensemble-review.js +371 -0
  496. package/fable-mode/scripts/model-selector.js +260 -0
  497. package/fable-mode/scripts/workflow-plan-consumer.js +337 -0
  498. package/find-skills/SKILL.md +48 -0
  499. package/find-skills/references/discovery-flow.md +102 -0
  500. package/find-skills/scripts/use-skill.js +112 -0
  501. package/git-commit/SKILL.md +28 -45
  502. package/git-commit/guides/ANGULAR_STYLE.md +60 -60
  503. package/git-commit/guides/COMMIT_GENERATION.md +23 -23
  504. package/git-commit/guides/LANGUAGE_DETECTION.md +16 -16
  505. package/git-commit/guides/MAIN_REPO.md +14 -14
  506. package/git-commit/guides/SUBMODULES.md +27 -27
  507. package/git-commit/references/commit-flow.md +30 -0
  508. package/grill-me/SKILL.md +34 -27
  509. package/grill-me/references/grilling-playbook.md +38 -0
  510. package/grill-with-docs/ADR-FORMAT.md +47 -47
  511. package/grill-with-docs/CONTEXT-FORMAT.md +63 -63
  512. package/grill-with-docs/SKILL.md +33 -91
  513. package/grill-with-docs/references/session-playbook.md +99 -0
  514. package/harness-everything/SKILL.md +27 -125
  515. package/harness-everything/examples/README.md +9 -9
  516. package/harness-everything/examples/todo-workflow.md +22 -22
  517. package/harness-everything/references/router-workflow-plan.md +128 -0
  518. package/harness-everything/references/skill-registry.md +39 -0
  519. package/harness-everything/references/triage-and-tiers.md +193 -0
  520. package/harness-everything/schemas/router-task-shape.schema.json +126 -0
  521. package/harness-everything/schemas/router-workflow-plan.schema.json +149 -0
  522. package/harness-everything/scripts/bootstrap.js +212 -140
  523. package/harness-everything/scripts/ensemble-policy.js +102 -0
  524. package/harness-everything/scripts/kernel-router.js +212 -0
  525. package/harness-everything/scripts/router-contract.js +612 -0
  526. package/harness-everything/scripts/routing-keywords.json +320 -0
  527. package/harness-everything/scripts/self-heal.js +250 -189
  528. package/harness-everything/scripts/tier-router.js +459 -219
  529. package/harness-everything/scripts/verify-gate.js +100 -31
  530. package/hooks/hooks.json +200 -155
  531. package/hooks/scripts/action-gate-rules.json +77 -0
  532. package/hooks/scripts/action-gate.js +528 -0
  533. package/hooks/scripts/atomic-commit-check.js +68 -68
  534. package/hooks/scripts/boundary-guard.js +55 -55
  535. package/hooks/scripts/codex-action-gate-post.js +46 -0
  536. package/hooks/scripts/codex-permission-request.js +140 -0
  537. package/hooks/scripts/context-compact.js +115 -114
  538. package/hooks/scripts/contract-test.js +146 -79
  539. package/hooks/scripts/depth-guard.js +65 -65
  540. package/hooks/scripts/lib/fable-contracts.js +121 -0
  541. package/hooks/scripts/lib/harness-state.js +275 -228
  542. package/hooks/scripts/lib/platforms/claude.js +97 -61
  543. package/hooks/scripts/lib/platforms/codex.js +42 -50
  544. package/hooks/scripts/lib/platforms/continue.js +30 -12
  545. package/hooks/scripts/lib/platforms/copilot.js +36 -42
  546. package/hooks/scripts/lib/platforms/cursor.js +23 -5
  547. package/hooks/scripts/lib/platforms/hermes.js +108 -57
  548. package/hooks/scripts/lib/platforms/index.js +17 -17
  549. package/hooks/scripts/lib/platforms/worktrees.js +20 -20
  550. package/hooks/scripts/openai-session-start.js +81 -0
  551. package/hooks/scripts/reset-circuit-breaker.js +30 -30
  552. package/hooks/scripts/rule-of-3-tracker.js +125 -91
  553. package/hooks/scripts/rule-of-3.js +195 -176
  554. package/hooks/scripts/state-persist.js +100 -99
  555. package/hooks/scripts/stop-gate.js +80 -80
  556. package/hooks/scripts/subagent-scope-guard.js +194 -88
  557. package/hooks/scripts/subagent-start.js +19 -0
  558. package/hooks/scripts/subagent-stop.js +6 -0
  559. package/improve-codebase-architecture/SKILL.md +29 -35
  560. package/improve-codebase-architecture/guides/DEEPENING.md +37 -37
  561. package/improve-codebase-architecture/guides/HTML-REPORT.md +123 -123
  562. package/improve-codebase-architecture/guides/INTERFACE-DESIGN.md +44 -44
  563. package/improve-codebase-architecture/guides/LANGUAGE.md +53 -53
  564. package/improve-codebase-architecture/guides/refactoring-workflow.md +59 -0
  565. package/improve-codebase-architecture/references/README.md +16 -0
  566. package/install-cognitive-os/SKILL.md +30 -81
  567. package/install-cognitive-os/references/cognitive-loop.md +27 -0
  568. package/multi-agent-workspace/NOTICE-agency-agents.md +27 -0
  569. package/multi-agent-workspace/SKILL.md +38 -0
  570. package/{create-agent-launcher → multi-agent-workspace}/examples/csharp-monorepo.md +123 -123
  571. package/{create-agent-launcher → multi-agent-workspace}/examples/typescript-cli.md +126 -126
  572. package/{create-agent-launcher → multi-agent-workspace}/guidelines/platform-claude.md +77 -77
  573. package/{create-agent-launcher → multi-agent-workspace}/guidelines/platform-codex.md +146 -146
  574. package/multi-agent-workspace/guidelines/platform-continue.md +38 -0
  575. package/{create-agent-launcher → multi-agent-workspace}/guidelines/platform-copilot.md +74 -74
  576. package/multi-agent-workspace/guidelines/platform-cursor.md +40 -0
  577. package/multi-agent-workspace/guidelines/platform-hermes.md +38 -0
  578. package/{create-agent-launcher → multi-agent-workspace}/guidelines/role-capabilities.md +108 -108
  579. package/multi-agent-workspace/references/agency-agents.md +31 -0
  580. package/multi-agent-workspace/references/architecture-guide.md +36 -0
  581. package/multi-agent-workspace/references/orchestration.md +65 -0
  582. package/multi-agent-workspace/scripts/agency-catalog.js +190 -0
  583. package/multi-agent-workspace/scripts/consume-workflow-plan.js +129 -0
  584. package/multi-agent-workspace/scripts/index_memory.js +104 -0
  585. package/multi-agent-workspace/scripts/path-boundary.js +59 -0
  586. package/multi-agent-workspace/scripts/project-docs-resolver.js +357 -0
  587. package/multi-agent-workspace/scripts/scaffold.js +408 -0
  588. package/multi-agent-workspace/templates/AGENTS.md +20 -0
  589. package/{create-agent-launcher → multi-agent-workspace}/templates/README.md +17 -17
  590. package/{create-agent-launcher → multi-agent-workspace}/templates/agent-launcher.md +67 -67
  591. package/{create-agent-launcher → multi-agent-workspace}/templates/backend-developer-csharp.md +55 -55
  592. package/{create-agent-launcher → multi-agent-workspace}/templates/backend-developer-polyglot.md +61 -61
  593. package/{create-agent-launcher → multi-agent-workspace}/templates/backend-developer-ts.md +56 -56
  594. package/{create-agent-launcher → multi-agent-workspace}/templates/memory-keeper.md +36 -36
  595. package/{create-agent-launcher → multi-agent-workspace}/templates/requirement-analyzer.md +65 -65
  596. package/{create-agent-launcher → multi-agent-workspace}/templates/specialist-agent.md +34 -34
  597. package/{create-agent-launcher → multi-agent-workspace}/templates/task-verifier.md +55 -55
  598. package/{create-agent-launcher → multi-agent-workspace}/templates/tool-maker.md +32 -32
  599. package/multi-agent-workspace/workflows/01-init.md +22 -0
  600. package/multi-agent-workspace/workflows/02-analysis.md +23 -0
  601. package/multi-agent-workspace/workflows/03-generation.md +25 -0
  602. package/multi-agent-workspace/workflows/04-launcher.md +16 -0
  603. package/opencode-plugin/README.md +137 -0
  604. package/opencode-plugin/hooks/verify.js +190 -0
  605. package/opencode-plugin/index.mjs +547 -0
  606. package/opencode-plugin/plugin.json +9 -0
  607. package/package.json +63 -22
  608. package/plugins/harness-everything/.codex-plugin/plugin.json +39 -0
  609. package/plugins/harness-everything/hooks/hooks.json +197 -0
  610. package/plugins/harness-everything/hooks/scripts/action-gate-rules.json +77 -0
  611. package/plugins/harness-everything/hooks/scripts/action-gate.js +528 -0
  612. package/plugins/harness-everything/hooks/scripts/atomic-commit-check.js +68 -0
  613. package/plugins/harness-everything/hooks/scripts/boundary-guard.js +55 -0
  614. package/plugins/harness-everything/hooks/scripts/codex-action-gate-post.js +46 -0
  615. package/plugins/harness-everything/hooks/scripts/codex-permission-request.js +140 -0
  616. package/plugins/harness-everything/hooks/scripts/context-compact.js +115 -0
  617. package/plugins/harness-everything/hooks/scripts/contract-test.js +146 -0
  618. package/plugins/harness-everything/hooks/scripts/depth-guard.js +65 -0
  619. package/plugins/harness-everything/hooks/scripts/lib/fable-contracts.js +121 -0
  620. package/plugins/harness-everything/hooks/scripts/lib/harness-state.js +275 -0
  621. package/plugins/harness-everything/hooks/scripts/lib/platforms/claude.js +168 -0
  622. package/plugins/harness-everything/hooks/scripts/lib/platforms/codex.js +88 -0
  623. package/plugins/harness-everything/hooks/scripts/lib/platforms/continue.js +91 -0
  624. package/plugins/harness-everything/hooks/scripts/lib/platforms/copilot.js +85 -0
  625. package/plugins/harness-everything/hooks/scripts/lib/platforms/cursor.js +89 -0
  626. package/plugins/harness-everything/hooks/scripts/lib/platforms/hermes.js +108 -0
  627. package/plugins/harness-everything/hooks/scripts/lib/platforms/index.js +17 -0
  628. package/plugins/harness-everything/hooks/scripts/lib/platforms/worktrees.js +20 -0
  629. package/plugins/harness-everything/hooks/scripts/lib/workspace.js +290 -0
  630. package/plugins/harness-everything/hooks/scripts/openai-session-start.js +81 -0
  631. package/plugins/harness-everything/hooks/scripts/reset-circuit-breaker.js +30 -0
  632. package/plugins/harness-everything/hooks/scripts/rule-of-3-tracker.js +125 -0
  633. package/plugins/harness-everything/hooks/scripts/rule-of-3.js +195 -0
  634. package/plugins/harness-everything/hooks/scripts/state-persist.js +100 -0
  635. package/plugins/harness-everything/hooks/scripts/stop-gate.js +80 -0
  636. package/plugins/harness-everything/hooks/scripts/subagent-scope-guard.js +194 -0
  637. package/plugins/harness-everything/hooks/scripts/subagent-start.js +19 -0
  638. package/plugins/harness-everything/hooks/scripts/subagent-stop.js +6 -0
  639. package/plugins/harness-everything/hooks/session-start.js +81 -0
  640. package/plugins/harness-everything/plugin.json +43 -0
  641. package/plugins/harness-everything/scripts/lib/advisory-text.js +174 -0
  642. package/plugins/harness-everything/skills/environment-detection/SKILL.md +41 -0
  643. package/plugins/harness-everything/skills/environment-detection/references/shell-syntax-rules.md +76 -0
  644. package/plugins/harness-everything/skills/environment-detection/scripts/preflight.js +103 -0
  645. package/plugins/harness-everything/skills/eval-harness/SKILL.md +42 -0
  646. package/plugins/harness-everything/skills/eval-harness/references/scoring-rubric.md +68 -0
  647. package/plugins/harness-everything/skills/eval-harness/scripts/evaluate.js +79 -0
  648. package/plugins/harness-everything/skills/fable-discipline/SKILL.md +41 -0
  649. package/plugins/harness-everything/skills/fable-discipline/references/discipline-rules.md +22 -0
  650. package/plugins/harness-everything/skills/fable-mode/CONTRACT-FORMAT.md +165 -0
  651. package/plugins/harness-everything/skills/fable-mode/SKILL.md +39 -0
  652. package/plugins/harness-everything/skills/fable-mode/agents/fable-orchestrator.md +159 -0
  653. package/plugins/harness-everything/skills/fable-mode/agents/fable-verifier.md +32 -0
  654. package/plugins/harness-everything/skills/fable-mode/agents/fable-worker-haiku.md +52 -0
  655. package/plugins/harness-everything/skills/fable-mode/agents/fable-worker-sonnet.md +53 -0
  656. package/plugins/harness-everything/skills/fable-mode/execution-guardrails/SKILL.md +43 -0
  657. package/plugins/harness-everything/skills/fable-mode/execution-guardrails/references/guardrail-rules.md +56 -0
  658. package/plugins/harness-everything/skills/fable-mode/fable-haiku/SKILL.md +36 -0
  659. package/plugins/harness-everything/skills/fable-mode/fable-haiku/references/routing-notes.md +26 -0
  660. package/plugins/harness-everything/skills/fable-mode/fable-opus/SKILL.md +36 -0
  661. package/plugins/harness-everything/skills/fable-mode/fable-sonnet/SKILL.md +36 -0
  662. package/plugins/harness-everything/skills/fable-mode/model-matrix.json +32 -0
  663. package/plugins/harness-everything/skills/fable-mode/references/execution-phases.md +37 -0
  664. package/plugins/harness-everything/skills/fable-mode/references/model-matrix.md +39 -0
  665. package/plugins/harness-everything/skills/fable-mode/scripts/ensemble-review.js +371 -0
  666. package/plugins/harness-everything/skills/fable-mode/scripts/model-selector.js +260 -0
  667. package/plugins/harness-everything/skills/fable-mode/scripts/workflow-plan-consumer.js +337 -0
  668. package/plugins/harness-everything/skills/find-skills/SKILL.md +48 -0
  669. package/plugins/harness-everything/skills/find-skills/references/discovery-flow.md +102 -0
  670. package/plugins/harness-everything/skills/find-skills/scripts/use-skill.js +112 -0
  671. package/plugins/harness-everything/skills/git-commit/SKILL.md +38 -0
  672. package/plugins/harness-everything/skills/git-commit/guides/ANGULAR_STYLE.md +61 -0
  673. package/plugins/harness-everything/skills/git-commit/guides/COMMIT_GENERATION.md +24 -0
  674. package/plugins/harness-everything/skills/git-commit/guides/LANGUAGE_DETECTION.md +16 -0
  675. package/plugins/harness-everything/skills/git-commit/guides/MAIN_REPO.md +15 -0
  676. package/plugins/harness-everything/skills/git-commit/guides/SUBMODULES.md +27 -0
  677. package/plugins/harness-everything/skills/git-commit/references/commit-flow.md +30 -0
  678. package/plugins/harness-everything/skills/grill-me/SKILL.md +43 -0
  679. package/plugins/harness-everything/skills/grill-me/references/grilling-playbook.md +38 -0
  680. package/plugins/harness-everything/skills/grill-with-docs/ADR-FORMAT.md +47 -0
  681. package/plugins/harness-everything/skills/grill-with-docs/CONTEXT-FORMAT.md +63 -0
  682. package/plugins/harness-everything/skills/grill-with-docs/SKILL.md +41 -0
  683. package/plugins/harness-everything/skills/grill-with-docs/references/session-playbook.md +99 -0
  684. package/plugins/harness-everything/skills/harness-everything/SKILL.md +43 -0
  685. package/plugins/harness-everything/skills/harness-everything/examples/README.md +9 -0
  686. package/plugins/harness-everything/skills/harness-everything/examples/todo-workflow.md +22 -0
  687. package/plugins/harness-everything/skills/harness-everything/references/router-workflow-plan.md +128 -0
  688. package/plugins/harness-everything/skills/harness-everything/references/skill-registry.md +39 -0
  689. package/plugins/harness-everything/skills/harness-everything/references/triage-and-tiers.md +193 -0
  690. package/plugins/harness-everything/skills/harness-everything/schemas/router-task-shape.schema.json +126 -0
  691. package/plugins/harness-everything/skills/harness-everything/schemas/router-workflow-plan.schema.json +149 -0
  692. package/plugins/harness-everything/skills/harness-everything/scripts/bootstrap.js +212 -0
  693. package/plugins/harness-everything/skills/harness-everything/scripts/ensemble-policy.js +102 -0
  694. package/plugins/harness-everything/skills/harness-everything/scripts/kernel-router.js +212 -0
  695. package/plugins/harness-everything/skills/harness-everything/scripts/router-contract.js +612 -0
  696. package/plugins/harness-everything/skills/harness-everything/scripts/routing-keywords.json +320 -0
  697. package/plugins/harness-everything/skills/harness-everything/scripts/self-heal.js +250 -0
  698. package/plugins/harness-everything/skills/harness-everything/scripts/tier-router.js +459 -0
  699. package/plugins/harness-everything/skills/harness-everything/scripts/verify-gate.js +100 -0
  700. package/plugins/harness-everything/skills/improve-codebase-architecture/SKILL.md +38 -0
  701. package/plugins/harness-everything/skills/improve-codebase-architecture/guides/DEEPENING.md +37 -0
  702. package/plugins/harness-everything/skills/improve-codebase-architecture/guides/HTML-REPORT.md +123 -0
  703. package/plugins/harness-everything/skills/improve-codebase-architecture/guides/INTERFACE-DESIGN.md +44 -0
  704. package/plugins/harness-everything/skills/improve-codebase-architecture/guides/LANGUAGE.md +53 -0
  705. package/plugins/harness-everything/skills/improve-codebase-architecture/guides/refactoring-workflow.md +59 -0
  706. package/plugins/harness-everything/skills/improve-codebase-architecture/references/README.md +16 -0
  707. package/plugins/harness-everything/skills/install-cognitive-os/SKILL.md +44 -0
  708. package/plugins/harness-everything/skills/install-cognitive-os/references/cognitive-loop.md +27 -0
  709. package/plugins/harness-everything/skills/multi-agent-workspace/NOTICE-agency-agents.md +27 -0
  710. package/plugins/harness-everything/skills/multi-agent-workspace/SKILL.md +38 -0
  711. package/plugins/harness-everything/skills/multi-agent-workspace/examples/csharp-monorepo.md +123 -0
  712. package/plugins/harness-everything/skills/multi-agent-workspace/examples/typescript-cli.md +126 -0
  713. package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-claude.md +77 -0
  714. package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-codex.md +146 -0
  715. package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-continue.md +38 -0
  716. package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-copilot.md +74 -0
  717. package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-cursor.md +40 -0
  718. package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-hermes.md +38 -0
  719. package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/role-capabilities.md +108 -0
  720. package/plugins/harness-everything/skills/multi-agent-workspace/references/agency-agents.md +31 -0
  721. package/plugins/harness-everything/skills/multi-agent-workspace/references/architecture-guide.md +36 -0
  722. package/plugins/harness-everything/skills/multi-agent-workspace/references/orchestration.md +65 -0
  723. package/plugins/harness-everything/skills/multi-agent-workspace/scripts/agency-catalog.js +190 -0
  724. package/plugins/harness-everything/skills/multi-agent-workspace/scripts/consume-workflow-plan.js +129 -0
  725. package/plugins/harness-everything/skills/multi-agent-workspace/scripts/index_memory.js +104 -0
  726. package/plugins/harness-everything/skills/multi-agent-workspace/scripts/path-boundary.js +59 -0
  727. package/plugins/harness-everything/skills/multi-agent-workspace/scripts/project-docs-resolver.js +357 -0
  728. package/plugins/harness-everything/skills/multi-agent-workspace/scripts/scaffold.js +408 -0
  729. package/plugins/harness-everything/skills/multi-agent-workspace/templates/AGENTS.md +20 -0
  730. package/plugins/harness-everything/skills/multi-agent-workspace/templates/README.md +17 -0
  731. package/plugins/harness-everything/skills/multi-agent-workspace/templates/agent-launcher.md +68 -0
  732. package/plugins/harness-everything/skills/multi-agent-workspace/templates/backend-developer-csharp.md +56 -0
  733. package/plugins/harness-everything/skills/multi-agent-workspace/templates/backend-developer-polyglot.md +62 -0
  734. package/plugins/harness-everything/skills/multi-agent-workspace/templates/backend-developer-ts.md +57 -0
  735. package/plugins/harness-everything/skills/multi-agent-workspace/templates/memory-keeper.md +37 -0
  736. package/plugins/harness-everything/skills/multi-agent-workspace/templates/requirement-analyzer.md +66 -0
  737. package/plugins/harness-everything/skills/multi-agent-workspace/templates/specialist-agent.md +34 -0
  738. package/plugins/harness-everything/skills/multi-agent-workspace/templates/task-verifier.md +56 -0
  739. package/plugins/harness-everything/skills/multi-agent-workspace/templates/tool-maker.md +33 -0
  740. package/plugins/harness-everything/skills/multi-agent-workspace/workflows/01-init.md +22 -0
  741. package/plugins/harness-everything/skills/multi-agent-workspace/workflows/02-analysis.md +23 -0
  742. package/plugins/harness-everything/skills/multi-agent-workspace/workflows/03-generation.md +25 -0
  743. package/plugins/harness-everything/skills/multi-agent-workspace/workflows/04-launcher.md +16 -0
  744. package/plugins/harness-everything/skills/repo-docs/SKILL.md +42 -0
  745. package/plugins/harness-everything/skills/repo-docs/references/process-guide.md +60 -0
  746. package/plugins/harness-everything/skills/repo-docs/templates/agents-template.md +264 -0
  747. package/plugins/harness-everything/skills/repo-docs/templates/knowledge-agents-template.md +50 -0
  748. package/plugins/harness-everything/skills/repo-docs/templates/knowledge-readme-template.md +83 -0
  749. package/plugins/harness-everything/skills/repo-docs/templates/manual-recon.md +117 -0
  750. package/plugins/harness-everything/skills/repo-docs/templates/multi-skills-readme-template.md +72 -0
  751. package/plugins/harness-everything/skills/repo-docs/templates/product-agents-template.md +78 -0
  752. package/plugins/harness-everything/skills/repo-docs/templates/product-readme-template.md +95 -0
  753. package/plugins/harness-everything/skills/repo-docs/templates/readme-template.md +136 -0
  754. package/plugins/harness-everything/skills/repo-docs/templates/single-skill-readme-template.md +70 -0
  755. package/plugins/harness-everything/skills/repo-docs/templates/skills-agents-template.md +60 -0
  756. package/plugins/harness-everything/skills/rewrite-commits/SKILL.md +37 -0
  757. package/plugins/harness-everything/skills/rewrite-commits/references/workflow.md +18 -0
  758. package/plugins/harness-everything/skills/security-review/SKILL.md +47 -0
  759. package/plugins/harness-everything/skills/security-review/guides/OWASP-PATTERNS.md +31 -0
  760. package/plugins/harness-everything/skills/security-review/guides/STRIDE-THREAT-MODEL.md +21 -0
  761. package/plugins/harness-everything/skills/security-review/references/security-checklist.md +478 -0
  762. package/plugins/harness-everything/skills/security-review/scripts/audit-secrets.js +85 -0
  763. package/plugins/harness-everything/skills/self-evolve/SKILL.md +45 -0
  764. package/plugins/harness-everything/skills/self-evolve/references/memory-resolution.md +61 -0
  765. package/plugins/harness-everything/skills/self-evolve/scripts/persist-memory.js +272 -0
  766. package/plugins/harness-everything/skills/self-evolve/scripts/register-dynamic-skill.js +253 -0
  767. package/plugins/harness-everything/skills/self-evolve/scripts/self-regression.js +125 -0
  768. package/plugins/harness-everything/skills/skill-creator/SKILL.md +47 -0
  769. package/plugins/harness-everything/skills/skill-creator/references/authoring-workflow.md +14 -0
  770. package/plugins/harness-everything/skills/skill-creator/references/dynamic-generation-contract.md +21 -0
  771. package/plugins/harness-everything/skills/skill-creator/references/quality-checklist.md +16 -0
  772. package/plugins/harness-everything/skills/skill-creator/references/quality-principles.md +81 -0
  773. package/plugins/harness-everything/skills/skill-creator/references/testing-workflow.md +27 -0
  774. package/plugins/harness-everything/skills/skill-style/SKILL.md +42 -0
  775. package/plugins/harness-everything/skills/skill-style/references/style-guide.md +62 -0
  776. package/plugins/harness-everything/skills/tdd/SKILL.md +38 -0
  777. package/plugins/harness-everything/skills/tdd/fixtures/pass-evidence.json +100 -0
  778. package/plugins/harness-everything/skills/tdd/fixtures/quality-scenarios.json +20 -0
  779. package/plugins/harness-everything/skills/tdd/guides/deep-modules.md +33 -0
  780. package/plugins/harness-everything/skills/tdd/guides/interface-design.md +31 -0
  781. package/plugins/harness-everything/skills/tdd/guides/mocking.md +59 -0
  782. package/plugins/harness-everything/skills/tdd/guides/refactoring.md +10 -0
  783. package/plugins/harness-everything/skills/tdd/guides/tests.md +61 -0
  784. package/plugins/harness-everything/skills/tdd/references/core-discipline.md +51 -0
  785. package/plugins/harness-everything/skills/tdd/references/integration-testing.md +13 -0
  786. package/plugins/harness-everything/skills/tdd/references/quality-model.md +68 -0
  787. package/plugins/harness-everything/skills/tdd/references/unit-testing.md +15 -0
  788. package/plugins/harness-everything/skills/tdd/schemas/quality-evidence.schema.json +211 -0
  789. package/plugins/harness-everything/skills/tdd/schemas/quality-report.schema.json +110 -0
  790. package/plugins/harness-everything/skills/tdd/scripts/quality-gate.js +475 -0
  791. package/plugins/harness-everything/skills/to-spec/SKILL.md +40 -0
  792. package/plugins/harness-everything/skills/to-spec/references/process.md +77 -0
  793. package/plugins/harness-everything/skills/to-spec/scripts/check-project-docs.js +239 -0
  794. package/plugins/harness-everything/skills/to-spec/scripts/project-docs-resolver.js +111 -0
  795. package/plugins/harness-everything/skills/to-spec/templates/cli-reference.md +31 -0
  796. package/plugins/harness-everything/skills/to-spec/templates/dev-doc.md +27 -0
  797. package/plugins/harness-everything/skills/to-spec/templates/feature-spec.md +59 -0
  798. package/plugins/harness-everything/skills/to-spec/templates/schema-doc.md +31 -0
  799. package/plugins/harness-everything/skills/to-tickets/SKILL.md +42 -0
  800. package/plugins/harness-everything/skills/to-tickets/references/publishing-and-templates.md +97 -0
  801. package/plugins/harness-everything/skills/todo-driven-workflow/SKILL.md +42 -0
  802. package/plugins/harness-everything/skills/todo-driven-workflow/references/execution-guide.md +27 -0
  803. package/plugins/harness-everything/skills/using-git-worktrees/SKILL.md +49 -0
  804. package/plugins/harness-everything/skills/using-git-worktrees/references/workflow-details.md +108 -0
  805. package/plugins/harness-everything/skills/verification-loop/SKILL.md +46 -0
  806. package/plugins/harness-everything/skills/verification-loop/references/verification-phases.md +116 -0
  807. package/plugins/harness-everything/skills/verification-loop/templates/verification-report.template.md +15 -0
  808. package/plugins/harness-everything/skills/verify-before-claim/SKILL.md +43 -0
  809. package/plugins/harness-everything/skills/verify-before-claim/references/verification-guide.md +73 -0
  810. package/plugins/harness-everything/skills/zoom-out/SKILL.md +42 -0
  811. package/plugins/harness-everything/skills/zoom-out/references/circuit-breaker.md +66 -0
  812. package/plugins/harness-everything/skills/zoom-out/templates/zoom-out-report.template.md +20 -0
  813. package/references/definition-of-done.md +67 -0
  814. package/references/performance-checklist.md +153 -0
  815. package/references/platform-advisory-template.md +58 -0
  816. package/references/security-checklist.md +205 -0
  817. package/repo-docs/SKILL.md +34 -37
  818. package/repo-docs/references/process-guide.md +60 -0
  819. package/repo-docs/templates/agents-template.md +264 -264
  820. package/repo-docs/templates/knowledge-agents-template.md +50 -50
  821. package/repo-docs/templates/knowledge-readme-template.md +83 -83
  822. package/repo-docs/templates/manual-recon.md +117 -117
  823. package/repo-docs/templates/multi-skills-readme-template.md +72 -72
  824. package/repo-docs/templates/product-agents-template.md +78 -78
  825. package/repo-docs/templates/product-readme-template.md +95 -95
  826. package/repo-docs/templates/readme-template.md +136 -136
  827. package/repo-docs/templates/single-skill-readme-template.md +70 -70
  828. package/repo-docs/templates/skills-agents-template.md +60 -60
  829. package/rewrite-commits/SKILL.md +22 -17
  830. package/rewrite-commits/references/workflow.md +18 -0
  831. package/scripts/build-openai-submission.js +158 -0
  832. package/scripts/installer.js +621 -506
  833. package/scripts/lib/advisory-text.js +33 -22
  834. package/scripts/lib/claude-hooks.js +18 -5
  835. package/scripts/lib/execution-contract.js +70 -0
  836. package/scripts/lib/gitignore.js +199 -80
  837. package/scripts/lib/manifest.js +83 -23
  838. package/scripts/lib/path-boundary.js +90 -0
  839. package/scripts/lib/prompts.js +1 -1
  840. package/scripts/lib/skills.js +191 -51
  841. package/scripts/lib/workspace.js +269 -7
  842. package/scripts/sync-openai-plugin.js +91 -0
  843. package/scripts/sync-release-version.js +212 -0
  844. package/scripts/verify-install.js +147 -0
  845. package/security-review/SKILL.md +33 -497
  846. package/security-review/guides/OWASP-PATTERNS.md +31 -0
  847. package/security-review/guides/STRIDE-THREAT-MODEL.md +21 -0
  848. package/security-review/references/security-checklist.md +478 -0
  849. package/security-review/scripts/audit-secrets.js +85 -0
  850. package/self-evolve/SKILL.md +36 -45
  851. package/self-evolve/references/memory-resolution.md +61 -0
  852. package/self-evolve/scripts/persist-memory.js +272 -40
  853. package/self-evolve/scripts/register-dynamic-skill.js +253 -0
  854. package/self-evolve/scripts/self-regression.js +125 -113
  855. package/skill-creator/SKILL.md +33 -77
  856. package/skill-creator/references/authoring-workflow.md +14 -0
  857. package/skill-creator/references/dynamic-generation-contract.md +21 -0
  858. package/skill-creator/references/quality-checklist.md +16 -0
  859. package/skill-creator/references/quality-principles.md +81 -81
  860. package/skill-creator/references/testing-workflow.md +27 -32
  861. package/skill-style/SKILL.md +33 -39
  862. package/skill-style/references/style-guide.md +62 -0
  863. package/submission/openai/README.md +67 -0
  864. package/submission/openai/listing.json +31 -0
  865. package/submission/openai/test-cases.json +61 -0
  866. package/tdd/SKILL.md +30 -45
  867. package/tdd/fixtures/pass-evidence.json +100 -0
  868. package/tdd/fixtures/quality-scenarios.json +20 -0
  869. package/tdd/guides/deep-modules.md +33 -33
  870. package/tdd/guides/interface-design.md +31 -31
  871. package/tdd/guides/mocking.md +59 -59
  872. package/tdd/guides/refactoring.md +10 -10
  873. package/tdd/guides/tests.md +61 -61
  874. package/tdd/references/core-discipline.md +51 -0
  875. package/tdd/references/integration-testing.md +13 -0
  876. package/tdd/references/quality-model.md +68 -0
  877. package/tdd/references/unit-testing.md +15 -0
  878. package/tdd/schemas/quality-evidence.schema.json +211 -0
  879. package/tdd/schemas/quality-report.schema.json +110 -0
  880. package/tdd/scripts/quality-gate.js +475 -0
  881. package/to-spec/SKILL.md +40 -0
  882. package/to-spec/references/process.md +77 -0
  883. package/to-spec/scripts/check-project-docs.js +239 -0
  884. package/to-spec/scripts/project-docs-resolver.js +111 -0
  885. package/to-spec/templates/cli-reference.md +31 -0
  886. package/to-spec/templates/dev-doc.md +27 -0
  887. package/to-spec/templates/feature-spec.md +59 -0
  888. package/to-spec/templates/schema-doc.md +31 -0
  889. package/to-tickets/SKILL.md +42 -0
  890. package/to-tickets/references/publishing-and-templates.md +97 -0
  891. package/todo-driven-workflow/SKILL.md +28 -40
  892. package/todo-driven-workflow/references/execution-guide.md +27 -0
  893. package/using-git-worktrees/SKILL.md +29 -141
  894. package/using-git-worktrees/references/workflow-details.md +108 -0
  895. package/verification-loop/SKILL.md +29 -121
  896. package/verification-loop/references/verification-phases.md +116 -0
  897. package/verification-loop/templates/verification-report.template.md +15 -0
  898. package/verify-before-claim/SKILL.md +27 -70
  899. package/verify-before-claim/references/verification-guide.md +73 -0
  900. package/zoom-out/SKILL.md +28 -76
  901. package/zoom-out/references/circuit-breaker.md +66 -0
  902. package/zoom-out/templates/zoom-out-report.template.md +20 -0
  903. package/build-multi-agent-system/AGENTS.md +0 -50
  904. package/build-multi-agent-system/README.md +0 -63
  905. package/build-multi-agent-system/SKILL.md +0 -68
  906. package/create-agent-launcher/SKILL.md +0 -48
  907. package/create-agent-launcher/guidelines/platform-gemini.md +0 -101
  908. package/create-agent-launcher/workflows/01-init.md +0 -22
  909. package/create-agent-launcher/workflows/02-analysis.md +0 -26
  910. package/create-agent-launcher/workflows/03-generation.md +0 -29
  911. package/create-agent-launcher/workflows/04-launcher.md +0 -20
  912. package/docs/workflows/build-multi-agent-system.md +0 -62
  913. package/docs/workflows/create-agent-launcher.md +0 -62
  914. package/eval-framework/behavioral-test.js +0 -105
  915. package/eval-framework/cases/case2.json +0 -5
  916. package/eval-framework/mechanism-test.js +0 -251
  917. package/harness-everything/scripts/todo-cli.js +0 -157
  918. package/skill-style/STYLE.md +0 -37
@@ -0,0 +1,35 @@
1
+ const fs = require('fs');
2
+ const path = require('path');
3
+ const helper = require('./test-helper');
4
+
5
+ console.log('\n[2u] Issue #44 release metadata regression checks...');
6
+
7
+ const packageJson = JSON.parse(fs.readFileSync(path.join(helper.root, 'package.json'), 'utf8'));
8
+ const skillText = name => fs.readFileSync(path.join(helper.root, name, 'SKILL.md'), 'utf8');
9
+ const versionOf = text => {
10
+ const match = text.match(/^ version:\s*(\S+)$/m);
11
+ return match && match[1];
12
+ };
13
+
14
+ helper.check(
15
+ '2u. modified fable-mode skill uses the released package version',
16
+ versionOf(skillText('fable-mode')) === packageJson.version,
17
+ `fable-mode=${versionOf(skillText('fable-mode'))}, package=${packageJson.version}`
18
+ );
19
+ for (const nested of ['execution-guardrails', 'fable-haiku', 'fable-opus', 'fable-sonnet']) {
20
+ helper.check(
21
+ `2u. fable-mode/${nested} inherits the parent version`,
22
+ versionOf(skillText(path.join('fable-mode', nested))) === versionOf(skillText('fable-mode')),
23
+ `${nested}=${versionOf(skillText(path.join('fable-mode', nested)))}, parent=${versionOf(skillText('fable-mode'))}`
24
+ );
25
+ }
26
+ helper.check(
27
+ '2u. modified todo-driven-workflow skill uses the released package version',
28
+ versionOf(skillText('todo-driven-workflow')) === packageJson.version,
29
+ `todo-driven-workflow=${versionOf(skillText('todo-driven-workflow'))}, package=${packageJson.version}`
30
+ );
31
+
32
+ const changelog = fs.readFileSync(path.join(helper.root, 'CHANGELOG.md'), 'utf8');
33
+ helper.check('2u. changelog records the Issue #44 disclosure policy', /#44/.test(changelog), 'CHANGELOG.md has no #44 entry');
34
+
35
+ helper.finish();
@@ -0,0 +1,138 @@
1
+ const fs = require('fs');
2
+ const os = require('os');
3
+ const path = require('path');
4
+ const helper = require('./test-helper');
5
+
6
+ const { assertContainedPath } = require('../scripts/lib/path-boundary');
7
+ const {
8
+ commandMatches,
9
+ commandsEqual,
10
+ extractPatchPaths,
11
+ patchPathsAuthorized,
12
+ } = require('../scripts/lib/execution-contract');
13
+
14
+ console.log('\n[2v] Verifier boundary contracts...');
15
+
16
+ const root = fs.mkdtempSync(path.join(os.tmpdir(), 'harness-2v-root-'));
17
+ const outside = fs.mkdtempSync(path.join(os.tmpdir(), 'harness-2v-outside-'));
18
+
19
+ try {
20
+ const workspace = path.join(root, 'workspace');
21
+ const nested = path.join(workspace, 'nested');
22
+ const outsideFile = path.join(outside, 'outside.txt');
23
+ fs.mkdirSync(nested, { recursive: true });
24
+ fs.writeFileSync(outsideFile, 'outside\n');
25
+
26
+ helper.check(
27
+ '2v. a normal nested path is inside its physical boundary',
28
+ assertContainedPath(workspace, path.join(nested, 'new.txt')).target.endsWith(path.join('nested', 'new.txt'))
29
+ );
30
+
31
+ helper.check(
32
+ '2v. lexical parent traversal is rejected',
33
+ (() => {
34
+ try {
35
+ assertContainedPath(workspace, path.join(workspace, '..', 'escape.txt'));
36
+ return false;
37
+ } catch (error) {
38
+ return error.code === 'PATH_OUTSIDE_BOUNDARY';
39
+ }
40
+ })()
41
+ );
42
+
43
+ const link = path.join(workspace, 'linked');
44
+ fs.symlinkSync(outside, link, process.platform === 'win32' ? 'junction' : 'dir');
45
+ helper.check(
46
+ '2v. a junction/symlink inside the boundary cannot escape it',
47
+ (() => {
48
+ try {
49
+ assertContainedPath(workspace, path.join(link, 'new.txt'));
50
+ return false;
51
+ } catch (error) {
52
+ return error.code === 'LINK_COMPONENT_IN_BOUNDARY';
53
+ }
54
+ })()
55
+ );
56
+
57
+ helper.check(
58
+ '2v. a missing leaf reached through a link is rejected before writing',
59
+ (() => {
60
+ try {
61
+ assertContainedPath(workspace, path.join(link, 'missing', 'new.txt'));
62
+ return false;
63
+ } catch (error) {
64
+ return error.code === 'LINK_COMPONENT_IN_BOUNDARY';
65
+ }
66
+ })()
67
+ );
68
+
69
+ const linkedInsideTarget = path.join(workspace, 'real-inside');
70
+ const linkedInside = path.join(workspace, 'linked-inside');
71
+ fs.mkdirSync(linkedInsideTarget, { recursive: true });
72
+ fs.symlinkSync(linkedInsideTarget, linkedInside, process.platform === 'win32' ? 'junction' : 'dir');
73
+ helper.check(
74
+ '2v. a link component is rejected even when its target stays inside the boundary',
75
+ (() => {
76
+ try {
77
+ assertContainedPath(workspace, path.join(linkedInside, 'new.txt'));
78
+ return false;
79
+ } catch (error) {
80
+ return error.code === 'LINK_COMPONENT_IN_BOUNDARY';
81
+ }
82
+ })()
83
+ );
84
+
85
+ const upperCaseWorkspace = workspace.toUpperCase();
86
+ helper.check(
87
+ '2v. Windows path casing does not change containment',
88
+ process.platform !== 'win32' || assertContainedPath(upperCaseWorkspace, path.join(workspace, 'nested', 'case.txt')).target
89
+ );
90
+
91
+ helper.check('2v. exact commands match after harmless whitespace normalization', commandsEqual(' npm test ', 'npm test'));
92
+ helper.check('2v. invalid boundary arguments use a stable error code', (() => {
93
+ try {
94
+ assertContainedPath(null, workspace);
95
+ return false;
96
+ } catch (error) {
97
+ return error.code === 'INVALID_BOUNDARY_PATH';
98
+ }
99
+ })());
100
+ helper.check('2v. a command containing the expected text is not an exact match', !commandsEqual('echo /tmp/preflight.js', 'preflight.js'));
101
+ helper.check('2v. extra command arguments are not silently accepted', !commandsEqual('npm test --coverage', 'npm test'));
102
+ helper.check('2v. argv commands compare each argument exactly', commandsEqual(['node', 'ci/test.js'], ['node', 'ci/test.js']) && !commandsEqual(['node', 'ci/test.js', '--verbose'], ['node', 'ci/test.js']));
103
+ helper.check('2v. a real node script invocation can use a basename expectation', commandMatches('node C:\\temp\\preflight.js', 'preflight.js'));
104
+ helper.check('2v. a non-launcher mention cannot satisfy a script expectation', !commandMatches('echo C:\\temp\\preflight.js', 'preflight.js'));
105
+
106
+ const patch = [
107
+ '*** Begin Patch',
108
+ '*** Update File: behavioral-evals/run.js',
109
+ '@@',
110
+ '*** Update File: ci/mechanism-2v-verifier-boundaries.test.js',
111
+ '@@',
112
+ '*** End Patch',
113
+ ].join('\n');
114
+ helper.check(
115
+ '2v. patch parser returns every touched path',
116
+ JSON.stringify(extractPatchPaths(patch)) === JSON.stringify([
117
+ 'behavioral-evals/run.js',
118
+ 'ci/mechanism-2v-verifier-boundaries.test.js',
119
+ ])
120
+ );
121
+ helper.check(
122
+ '2v. a multi-file patch is authorized only when every path is allowed',
123
+ patchPathsAuthorized(patch, new Set(['behavioral-evals/run.js', 'ci/mechanism-2v-verifier-boundaries.test.js']))
124
+ );
125
+ helper.check(
126
+ '2v. one unrelated path rejects the whole patch',
127
+ !patchPathsAuthorized(patch, new Set(['behavioral-evals/run.js']))
128
+ );
129
+ helper.check(
130
+ '2v. traversal paths are never authorized even when allowlisted',
131
+ !patchPathsAuthorized('*** Update File: ../escape.js\n', new Set(['../escape.js']))
132
+ );
133
+ } finally {
134
+ fs.rmSync(root, { recursive: true, force: true });
135
+ fs.rmSync(outside, { recursive: true, force: true });
136
+ }
137
+
138
+ helper.finish();
@@ -0,0 +1,148 @@
1
+ 'use strict';
2
+
3
+ const assert = require('assert');
4
+ const crypto = require('crypto');
5
+ const fs = require('fs');
6
+ const os = require('os');
7
+ const path = require('path');
8
+
9
+ const ROOT = path.resolve(__dirname, '..');
10
+ const matrix = JSON.parse(fs.readFileSync(path.join(ROOT, 'docs', 'platform-compatibility.json'), 'utf8'));
11
+ const platforms = require('../hooks/scripts/lib/platforms');
12
+ const manifest = require('../scripts/lib/manifest');
13
+ const { build } = require('../scripts/build-openai-submission');
14
+
15
+ const DIMENSIONS = [
16
+ 'standaloneSkills',
17
+ 'pluginSupport',
18
+ 'bundledSkills',
19
+ 'hooksLifecycle',
20
+ 'projectScope',
21
+ 'globalScope',
22
+ 'installUninstallSymmetry',
23
+ 'assetsReferences',
24
+ 'updateSync',
25
+ 'liveHostVerification'
26
+ ];
27
+ const STATUSES = new Set(['Supported', 'Mechanism verified', 'Live verified', 'Partial', 'Unsupported', 'Unknown']);
28
+ const OFFICIAL_HOSTS = {
29
+ codex: new Set(['learn.chatgpt.com']),
30
+ 'chatgpt-openai-plugin': new Set(['help.openai.com', 'developers.openai.com']),
31
+ 'claude-code': new Set(['code.claude.com']),
32
+ opencode: new Set(['opencode.ai']),
33
+ 'github-copilot': new Set(['docs.github.com']),
34
+ cursor: new Set(['cursor.com', 'prod.cursor.com']),
35
+ continue: new Set(['docs.continue.dev']),
36
+ hermes: new Set(['hermes-agent.nousresearch.com'])
37
+ };
38
+
39
+ function filesUnder(root, base = root, out = []) {
40
+ for (const entry of fs.readdirSync(root, { withFileTypes: true })) {
41
+ const full = path.join(root, entry.name);
42
+ if (entry.isDirectory()) filesUnder(full, base, out);
43
+ else if (entry.isFile()) out.push(path.relative(base, full).replace(/\\/g, '/'));
44
+ }
45
+ return out.sort();
46
+ }
47
+
48
+ function sha256(file) {
49
+ return crypto.createHash('sha256').update(fs.readFileSync(file)).digest('hex');
50
+ }
51
+
52
+ assert.strictEqual(matrix.format, 'harness-platform-compatibility-v1');
53
+ assert.deepStrictEqual(Object.keys(matrix.status_vocabulary).sort(), [...STATUSES].sort());
54
+ assert.deepStrictEqual(matrix.dimensions, DIMENSIONS);
55
+ assert.strictEqual(matrix.platforms.length, 8, 'compatibility matrix must cover every issue #82 platform');
56
+ assert.deepStrictEqual(
57
+ matrix.platforms.map(platform => platform.id).sort(),
58
+ Object.keys(OFFICIAL_HOSTS).sort()
59
+ );
60
+
61
+ for (const platform of matrix.platforms) {
62
+ assert.ok(Array.isArray(platform.official_sources) && platform.official_sources.length > 0, `${platform.id}: official source list missing`);
63
+ for (const source of platform.official_sources) {
64
+ const url = new URL(source);
65
+ assert.strictEqual(url.protocol, 'https:', `${platform.id}: source must use HTTPS`);
66
+ assert.ok(OFFICIAL_HOSTS[platform.id].has(url.hostname), `${platform.id}: source is not an approved official documentation host: ${source}`);
67
+ }
68
+ assert.deepStrictEqual(Object.keys(platform.capabilities).sort(), [...DIMENSIONS].sort(), `${platform.id}: dimension coverage drifted`);
69
+ for (const dimension of DIMENSIONS) {
70
+ const capability = platform.capabilities[dimension];
71
+ assert.ok(capability && STATUSES.has(capability.status), `${platform.id}/${dimension}: invalid status`);
72
+ }
73
+ assert.strictEqual(
74
+ platform.capabilities.liveHostVerification.status,
75
+ 'Unknown',
76
+ `${platform.id}: a live status requires a preserved host artifact before publication`
77
+ );
78
+ }
79
+
80
+ const installerTargetSpecs = {
81
+ codex: { platform: 'codex', project: '.agents/skills', global: '~/.agents/skills' },
82
+ 'claude-code': { platform: 'claude', project: '.claude/skills', global: '~/.claude/skills' },
83
+ cursor: { platform: 'cursor', project: '.cursor/skills', global: '~/.agents/skills' },
84
+ 'github-copilot': { platform: 'copilot', project: '.github/skills', global: '~/.agents/skills' },
85
+ continue: { platform: 'continue', project: '.continue/skills', global: '~/.continue/skills' },
86
+ hermes: { platform: 'hermes', project: '.agents/skills', global: '~/.hermes/skills' }
87
+ };
88
+ assert.deepStrictEqual(Object.keys(matrix.installer_targets).sort(), Object.keys(OFFICIAL_HOSTS).sort());
89
+ for (const [id, spec] of Object.entries(installerTargetSpecs)) {
90
+ const targetSpec = matrix.installer_targets[id];
91
+ assert.strictEqual(targetSpec.project, spec.project, `${id}: project installer target drifted`);
92
+ assert.strictEqual(targetSpec.global, spec.global, `${id}: global installer target drifted`);
93
+ assert.ok(STATUSES.has(targetSpec.host_discovery_status), `${id}: installer target status is invalid`);
94
+ }
95
+
96
+ const workspaceRoot = fs.mkdtempSync(path.join(os.tmpdir(), 'harness-2w-workspace-'));
97
+ const userHome = fs.mkdtempSync(path.join(os.tmpdir(), 'harness-2w-home-'));
98
+ try {
99
+ for (const [id, spec] of Object.entries(installerTargetSpecs)) {
100
+ const adapter = platforms.find(platform => platform.name === spec.platform);
101
+ assert.ok(adapter, `${id}: installer adapter missing`);
102
+ const local = adapter.getSkillsTarget({ workspaceRoot, userHome, isGlobal: false, manifest });
103
+ const global = adapter.getSkillsTarget({ workspaceRoot, userHome, isGlobal: true, manifest });
104
+ const expectedLocal = path.join(workspaceRoot, ...spec.project.split('/'));
105
+ const expectedGlobal = path.join(userHome, ...spec.global.replace('~/', '').split('/'));
106
+ assert.strictEqual(path.resolve(local.path), path.resolve(expectedLocal), `${id}: local adapter target drifted`);
107
+ assert.strictEqual(path.resolve(global.path), path.resolve(expectedGlobal), `${id}: global adapter target drifted`);
108
+ }
109
+ } finally {
110
+ fs.rmSync(workspaceRoot, { recursive: true, force: true });
111
+ fs.rmSync(userHome, { recursive: true, force: true });
112
+ }
113
+
114
+ const portableManifest = JSON.parse(fs.readFileSync(path.join(ROOT, 'plugins', 'harness-everything', 'plugin.json'), 'utf8'));
115
+ const compatibilityManifest = JSON.parse(fs.readFileSync(path.join(ROOT, 'plugins', 'harness-everything', '.codex-plugin', 'plugin.json'), 'utf8'));
116
+ assert.strictEqual(portableManifest.$schema, 'https://agent-plugins.org/schemas/1.0.0/plugin.schema.json');
117
+ assert.strictEqual(portableManifest.extensions?.['com.openai']?.hooks, './hooks/hooks.json');
118
+ assert.deepStrictEqual(portableManifest.extensions?.['com.openai']?.interface, compatibilityManifest.interface);
119
+ assert.ok(fs.existsSync(path.join(ROOT, 'plugins', 'harness-everything', 'skills', 'harness-everything', 'SKILL.md')));
120
+
121
+ const marketplace = JSON.parse(fs.readFileSync(path.join(ROOT, '.agents', 'plugins', 'marketplace.json'), 'utf8'));
122
+ const marketplaceEntry = marketplace.plugins.find(plugin => plugin.name === 'harness-everything');
123
+ assert.strictEqual(marketplaceEntry?.source?.path, './plugins/harness-everything');
124
+ assert.ok(fs.existsSync(path.join(ROOT, marketplaceEntry.source.path, '.codex-plugin', 'plugin.json')));
125
+ const claudeMarketplace = JSON.parse(fs.readFileSync(path.join(ROOT, '.claude-plugin', 'marketplace.json'), 'utf8'));
126
+ assert.ok(claudeMarketplace.plugins.some(plugin => plugin.source === './'), 'Claude marketplace must remain a repository-root import path');
127
+
128
+ const tempOutput = path.join(os.tmpdir(), `harness-2w-submission-${process.pid}.zip`);
129
+ try {
130
+ const result = build(tempOutput);
131
+ const publicManifest = JSON.parse(fs.readFileSync(result.manifestPath, 'utf8'));
132
+ const packagedFiles = filesUnder(path.join(ROOT, 'plugins', 'harness-everything', 'skills')).map(file => `skills/${file}`);
133
+ assert.deepStrictEqual(publicManifest.files.map(file => file.path), packagedFiles);
134
+ assert.ok(publicManifest.files.every(file => file.path.startsWith('skills/')));
135
+ assert.ok(publicManifest.files.every(file => !/(^|\/)(hooks|\.codex-plugin)(\/|$)/.test(file.path)));
136
+ assert.strictEqual(publicManifest.bundle_sha256, sha256(tempOutput));
137
+ } finally {
138
+ fs.rmSync(tempOutput, { force: true });
139
+ fs.rmSync(tempOutput.replace(/\.zip$/i, '') + '.manifest.json', { force: true });
140
+ }
141
+
142
+ const capabilityDocs = fs.readFileSync(path.join(ROOT, 'docs', 'platform-capabilities.md'), 'utf8');
143
+ assert.match(capabilityDocs, /platform-compatibility\.json/);
144
+ assert.match(capabilityDocs, /Continue\.dev[\s\S]*`Unknown` for standalone `SKILL\.md` discovery/);
145
+ assert.match(capabilityDocs, /live plugin loading remains unverified/i);
146
+ assert.match(capabilityDocs, /no fresh host-session artifacts/i);
147
+
148
+ console.log('Platform compatibility verified: official-source matrix, installer targets, package manifests, and public bundle boundaries.');
@@ -0,0 +1,157 @@
1
+ 'use strict';
2
+
3
+ const assert = require('assert');
4
+ const fs = require('fs');
5
+ const os = require('os');
6
+ const path = require('path');
7
+ const { execFileSync } = require('child_process');
8
+ const { VERSION_TARGETS, isStableSemVer, syncReleaseVersion } = require('../scripts/sync-release-version');
9
+
10
+ const ROOT = path.resolve(__dirname, '..');
11
+ const config = JSON.parse(fs.readFileSync(path.join(ROOT, '.releaserc.json'), 'utf8'));
12
+
13
+ assert.deepStrictEqual(config.branches, ['main']);
14
+ assert.ok(config.branches.every(branch => typeof branch === 'string'), 'release branches must not declare prerelease channels');
15
+ assert.strictEqual(config.tagFormat, 'v${version}');
16
+
17
+ const analyzer = config.plugins.find(plugin => Array.isArray(plugin) && plugin[0] === '@semantic-release/commit-analyzer');
18
+ assert.ok(analyzer, 'commit analyzer configuration missing');
19
+ const rules = new Map(analyzer[1].releaseRules.filter(rule => rule.type).map(rule => [rule.type, rule.release]));
20
+ for (const [type, release] of Object.entries({
21
+ feat: 'minor',
22
+ fix: 'patch',
23
+ perf: 'patch',
24
+ refactor: 'patch',
25
+ build: 'patch',
26
+ revert: 'patch'
27
+ })) {
28
+ assert.strictEqual(rules.get(type), release, `${type} release rule`);
29
+ }
30
+ for (const type of ['docs', 'test', 'ci', 'style', 'chore']) {
31
+ assert.ok(!rules.has(type), `${type} must fall through to semantic-release's no-release default`);
32
+ }
33
+ assert.ok(analyzer[1].releaseRules.some(rule => rule.breaking === true && rule.release === 'major'));
34
+
35
+ const execPlugin = config.plugins.find(plugin => Array.isArray(plugin) && plugin[0] === '@semantic-release/exec');
36
+ assert.ok(execPlugin, 'release preparation hook missing');
37
+ assert.match(execPlugin[1].verifyReleaseCmd, /--validate \$\{nextRelease\.version\}/);
38
+ assert.match(execPlugin[1].prepareCmd, /sync-release-version\.js \$\{nextRelease\.version\}/);
39
+ assert.match(execPlugin[1].prepareCmd, /--base \$\{lastRelease\.gitTag\}/, 'changed skills must be computed from the previous release tag');
40
+ assert.match(execPlugin[1].prepareCmd, /--check \$\{nextRelease\.version\} --base \$\{lastRelease\.gitTag\}/);
41
+
42
+ const gitPlugin = config.plugins.find(plugin => Array.isArray(plugin) && plugin[0] === '@semantic-release/git');
43
+ assert.ok(gitPlugin, 'release commit plugin missing');
44
+ assert.match(gitPlugin[1].message, /\[skip ci\]/);
45
+ assert.ok(gitPlugin[1].assets.includes('package-lock.json'));
46
+ assert.ok(gitPlugin[1].assets.includes('opencode-plugin/plugin.json'));
47
+ assert.ok(gitPlugin[1].assets.includes('*/SKILL.md'));
48
+ assert.ok(gitPlugin[1].assets.includes('*/**/SKILL.md'));
49
+
50
+ assert.ok(isStableSemVer('1.2.3'));
51
+ for (const invalid of ['1.2.3-beta', '1.2.3+build', '01.2.3', '1.2', 'v1.2.3', '']) {
52
+ assert.ok(!isStableSemVer(invalid), invalid);
53
+ }
54
+
55
+ // Normal CI gates every checked-in runtime/plugin manifest against package.json.
56
+ // package-lock is intentionally release-owned: the synchronizer fixes its root
57
+ // version when semantic-release prepares the actual release transaction.
58
+ const packageVersion = JSON.parse(fs.readFileSync(path.join(ROOT, 'package.json'), 'utf8')).version;
59
+ assert.ok(isStableSemVer(packageVersion), `package.json must use stable SemVer, got ${packageVersion}`);
60
+ for (const target of VERSION_TARGETS.filter(target => target.file !== 'package-lock.json')) {
61
+ const data = JSON.parse(fs.readFileSync(path.join(ROOT, target.file), 'utf8'));
62
+ const current = target.get(data);
63
+ const values = Array.isArray(current) ? current : [current];
64
+ assert.ok(values.every(value => value === packageVersion), `${target.file} must match package.json version ${packageVersion}`);
65
+ assert.ok(values.every(isStableSemVer), `${target.file} must not carry a prerelease/build suffix`);
66
+ }
67
+
68
+ function json(file, value) {
69
+ fs.mkdirSync(path.dirname(file), { recursive: true });
70
+ fs.writeFileSync(file, `${JSON.stringify(value, null, 2)}\n`);
71
+ }
72
+
73
+ function skill(file, name, version) {
74
+ fs.mkdirSync(path.dirname(file), { recursive: true });
75
+ fs.writeFileSync(file, `---\nname: ${name}\nmetadata:\n author: Test\n version: ${version}\n---\n\n# ${name}\n`);
76
+ }
77
+
78
+ function versionOfSkill(file) {
79
+ const match = fs.readFileSync(file, 'utf8').match(/^ version:\s*(\S+)$/m);
80
+ return match && match[1];
81
+ }
82
+
83
+ function fixture() {
84
+ const root = fs.mkdtempSync(path.join(os.tmpdir(), 'harness-release-'));
85
+ json(path.join(root, 'package.json'), { name: 'harness-everything', version: '0.3.8-beta' });
86
+ json(path.join(root, 'package-lock.json'), {
87
+ name: 'harness-everything',
88
+ version: '0.3.7-beta',
89
+ packages: { '': { name: 'harness-everything', version: '0.3.7-beta' } }
90
+ });
91
+ json(path.join(root, '.claude-plugin/plugin.json'), { name: 'harness-everything', version: '0.3.8-beta', skills: [] });
92
+ json(path.join(root, '.claude-plugin/marketplace.json'), { plugins: [{ name: 'harness-everything', version: '0.3.8-beta' }] });
93
+ json(path.join(root, 'plugins/harness-everything/plugin.json'), { name: 'harness-everything', version: '0.3.8-beta' });
94
+ json(path.join(root, 'plugins/harness-everything/.codex-plugin/plugin.json'), { name: 'harness-everything', version: '0.3.8-beta' });
95
+ json(path.join(root, 'opencode-plugin/plugin.json'), { name: 'harness-enforcement', version: '0.3.6' });
96
+ return root;
97
+ }
98
+
99
+ const root = fixture();
100
+ const files = syncReleaseVersion(root, '1.2.3');
101
+ assert.strictEqual(files.length, 7);
102
+ syncReleaseVersion(root, '1.2.3', { checkOnly: true });
103
+
104
+ const first = new Map(files.map(file => [file, fs.readFileSync(path.join(root, file), 'utf8')]));
105
+ syncReleaseVersion(root, '1.2.3');
106
+ for (const file of files) {
107
+ assert.strictEqual(fs.readFileSync(path.join(root, file), 'utf8'), first.get(file), `${file} must be idempotent`);
108
+ }
109
+
110
+ const lock = JSON.parse(fs.readFileSync(path.join(root, 'package-lock.json'), 'utf8'));
111
+ assert.strictEqual(lock.version, '1.2.3');
112
+ assert.strictEqual(lock.packages[''].version, '1.2.3');
113
+ const market = JSON.parse(fs.readFileSync(path.join(root, '.claude-plugin/marketplace.json'), 'utf8'));
114
+ assert.strictEqual(market.plugins[0].version, '1.2.3');
115
+
116
+ const portable = path.join(root, 'plugins/harness-everything/plugin.json');
117
+ const tampered = JSON.parse(fs.readFileSync(portable, 'utf8'));
118
+ tampered.version = '1.2.2';
119
+ json(portable, tampered);
120
+ assert.throws(() => syncReleaseVersion(root, '1.2.3', { checkOnly: true }), /Release version drift/);
121
+ assert.throws(() => syncReleaseVersion(root, '1.2.3-beta'), /stable SemVer/);
122
+
123
+ const missing = fixture();
124
+ const packageBefore = fs.readFileSync(path.join(missing, 'package.json'), 'utf8');
125
+ fs.rmSync(path.join(missing, 'opencode-plugin/plugin.json'));
126
+ assert.throws(() => syncReleaseVersion(missing, '2.0.0'), /Missing release version target/);
127
+ assert.strictEqual(fs.readFileSync(path.join(missing, 'package.json'), 'utf8'), packageBefore, 'validation failure must happen before writes');
128
+
129
+ // A release bumps only canonical skills changed since the previous release tag.
130
+ // Nested sub-skills inherit the parent version whenever any file under that
131
+ // top-level skill changed, while untouched skills keep their last-change version.
132
+ const changed = fixture();
133
+ const sourceManifest = JSON.parse(fs.readFileSync(path.join(changed, '.claude-plugin/plugin.json'), 'utf8'));
134
+ sourceManifest.skills = ['./alpha', './beta'];
135
+ json(path.join(changed, '.claude-plugin/plugin.json'), sourceManifest);
136
+ skill(path.join(changed, 'alpha/SKILL.md'), 'alpha', '0.1.0');
137
+ skill(path.join(changed, 'alpha/nested/SKILL.md'), 'nested', '0.1.0');
138
+ skill(path.join(changed, 'beta/SKILL.md'), 'beta', '0.1.0');
139
+ execFileSync('git', ['init'], { cwd: changed, stdio: 'ignore' });
140
+ execFileSync('git', ['config', 'user.email', 'release-test@example.com'], { cwd: changed });
141
+ execFileSync('git', ['config', 'user.name', 'Release Test'], { cwd: changed });
142
+ execFileSync('git', ['add', '.'], { cwd: changed });
143
+ execFileSync('git', ['commit', '-m', 'feat: baseline'], { cwd: changed, stdio: 'ignore' });
144
+ execFileSync('git', ['tag', 'v0.1.0'], { cwd: changed });
145
+ fs.writeFileSync(path.join(changed, 'alpha/behavior.txt'), 'changed\n');
146
+ execFileSync('git', ['add', '.'], { cwd: changed });
147
+ execFileSync('git', ['commit', '-m', 'feat(alpha): change behavior'], { cwd: changed, stdio: 'ignore' });
148
+ const changedFiles = syncReleaseVersion(changed, '0.2.0', { base: 'v0.1.0' });
149
+ assert.ok(changedFiles.includes('alpha/SKILL.md'));
150
+ assert.ok(changedFiles.includes('alpha/nested/SKILL.md'));
151
+ assert.ok(!changedFiles.includes('beta/SKILL.md'));
152
+ assert.strictEqual(versionOfSkill(path.join(changed, 'alpha/SKILL.md')), '0.2.0');
153
+ assert.strictEqual(versionOfSkill(path.join(changed, 'alpha/nested/SKILL.md')), '0.2.0');
154
+ assert.strictEqual(versionOfSkill(path.join(changed, 'beta/SKILL.md')), '0.1.0');
155
+ syncReleaseVersion(changed, '0.2.0', { base: 'v0.1.0', checkOnly: true });
156
+
157
+ console.log('Release automation contract verified: stable SemVer rules, changed-skill versioning, manifest sync, drift detection, idempotency, and fail-before-write behavior.');
@@ -0,0 +1,66 @@
1
+ 'use strict';
2
+
3
+ const assert = require('assert');
4
+ const fs = require('fs');
5
+ const os = require('os');
6
+ const path = require('path');
7
+ const { spawnSync } = require('child_process');
8
+
9
+ const ROOT = path.resolve(__dirname, '..');
10
+ const config = JSON.parse(fs.readFileSync(path.join(ROOT, '.releaserc.json'), 'utf8'));
11
+ const gitPlugin = config.plugins.find(plugin => Array.isArray(plugin) && plugin[0] === '@semantic-release/git');
12
+
13
+ assert.ok(gitPlugin, 'release commit plugin missing');
14
+ const template = gitPlugin[1].message;
15
+ assert.match(template, /\$\{nextRelease\.version\}/, 'release commit must include the calculated version');
16
+ assert.match(template, /\[skip ci\]/, 'release commit must not recursively trigger CI');
17
+ assert.ok(!template.includes('${nextRelease.notes}'), 'release notes must not be copied into the release commit message');
18
+
19
+ const currentMessage = template.replace('${nextRelease.version}', '0.4.0');
20
+ const currentLines = currentMessage.split(/\r?\n/);
21
+ assert.strictEqual(currentLines.length, 1, 'release commit must remain a single-line control message');
22
+ assert.ok(currentLines.every(line => line.length <= 100), 'release commit must remain within commitlint line limits');
23
+
24
+ const temp = fs.mkdtempSync(path.join(os.tmpdir(), 'harness-release-commitlint-'));
25
+ const currentFile = path.join(temp, 'current-message.txt');
26
+ const legacyFile = path.join(temp, 'legacy-message.txt');
27
+ fs.writeFileSync(currentFile, `${currentMessage}\n`);
28
+
29
+ // Negative control: this reproduces the production failure from release run
30
+ // 34970424994. Generated notes contain long Markdown links, so copying them
31
+ // into the Git commit body violates body-max-line-length/footer constraints.
32
+ const longReleaseNote = `* **behavioral-evals:** ${'x'.repeat(110)} ([90e36c2](https://github.com/dyphn1/Harness-everything/commit/90e36c2071fb5a71e16cdc99b1060cd3d699c43b)), closes #52`;
33
+ const legacyMessage = `chore(release): 0.4.0 [skip ci]\n\n${longReleaseNote}\n`;
34
+ fs.writeFileSync(legacyFile, legacyMessage);
35
+
36
+ const commitlintPackageFile = require.resolve('@commitlint/cli/package.json');
37
+ const commitlintPackage = JSON.parse(fs.readFileSync(commitlintPackageFile, 'utf8'));
38
+ const commitlintBin = typeof commitlintPackage.bin === 'string'
39
+ ? commitlintPackage.bin
40
+ : commitlintPackage.bin && commitlintPackage.bin.commitlint;
41
+ assert.ok(commitlintBin, '@commitlint/cli must expose a commitlint executable');
42
+ const commitlintCli = path.resolve(path.dirname(commitlintPackageFile), commitlintBin);
43
+
44
+ function commitlint(file) {
45
+ return spawnSync(process.execPath, [commitlintCli, '--edit', file], {
46
+ cwd: ROOT,
47
+ encoding: 'utf8'
48
+ });
49
+ }
50
+
51
+ const currentResult = commitlint(currentFile);
52
+ assert.strictEqual(
53
+ currentResult.status,
54
+ 0,
55
+ `current semantic-release commit must pass commitlint:\n${currentResult.stdout}\n${currentResult.stderr}`
56
+ );
57
+
58
+ const legacyResult = commitlint(legacyFile);
59
+ assert.notStrictEqual(legacyResult.status, 0, 'legacy release-notes-in-commit behavior must fail commitlint');
60
+ assert.match(
61
+ `${legacyResult.stdout}\n${legacyResult.stderr}`,
62
+ /body-max-line-length|footer-max-line-length/,
63
+ 'negative control must fail for the same long-line class seen in production'
64
+ );
65
+
66
+ console.log('Release commit contract verified: compact message passes commitlint and legacy release-note embedding is rejected.');
@@ -0,0 +1,47 @@
1
+ 'use strict';
2
+
3
+ const assert = require('assert');
4
+ const fs = require('fs');
5
+ const path = require('path');
6
+
7
+ const ROOT = path.resolve(__dirname, '..');
8
+ const workflow = fs.readFileSync(path.join(ROOT, '.github/workflows/release.yml'), 'utf8');
9
+
10
+ function pinnedVersion(packageName, source = workflow) {
11
+ const escaped = packageName.replace(/[.*+?^${}()|[\]\\]/g, '\\$&');
12
+ const match = source.match(new RegExp(`${escaped}@(\\d+\\.\\d+\\.\\d+(?:-[0-9A-Za-z.-]+)?)`));
13
+ assert.ok(match, `release workflow must pin ${packageName}`);
14
+ return match[1];
15
+ }
16
+
17
+ function major(version) {
18
+ return Number(version.split('.')[0]);
19
+ }
20
+
21
+ function assertCompatible(generatorVersion, presetVersion) {
22
+ if (major(generatorVersion) === 14 && major(presetVersion) >= 10) {
23
+ throw new Error(
24
+ `incompatible release-note tooling: @semantic-release/release-notes-generator@${generatorVersion} ` +
25
+ `uses conventional-changelog-writer 8, but conventional-changelog-conventionalcommits@${presetVersion} requires writer 9+`
26
+ );
27
+ }
28
+ }
29
+
30
+ const generatorVersion = pinnedVersion('@semantic-release/release-notes-generator');
31
+ const presetVersion = pinnedVersion('conventional-changelog-conventionalcommits');
32
+
33
+ assert.strictEqual(generatorVersion, '14.1.1', 'stable release-notes-generator pin changed; re-evaluate writer compatibility');
34
+ assert.strictEqual(presetVersion, '9.3.1', 'preset must stay on the writer-8-compatible line while generator 14 is used');
35
+ assert.doesNotThrow(() => assertCompatible(generatorVersion, presetVersion));
36
+ assert.throws(
37
+ () => assertCompatible('14.1.1', '10.4.0'),
38
+ /incompatible release-note tooling/,
39
+ 'negative control must reject the exact #113 failure pairing'
40
+ );
41
+ assert.doesNotMatch(
42
+ workflow,
43
+ /(?:^|\s)conventional-changelog-writer@\d/m,
44
+ 'do not force writer 9 into release-notes-generator 14; upgrade the generator as one compatible stack instead'
45
+ );
46
+
47
+ console.log('Release tooling compatibility verified: generator 14 is paired with conventionalcommits 9.x, and the v10 regression is rejected.');
@@ -0,0 +1,29 @@
1
+ #!/usr/bin/env node
2
+
3
+ const assert = require('assert');
4
+ const fs = require('fs');
5
+ const path = require('path');
6
+
7
+ const projectRoot = path.resolve(__dirname, '..');
8
+ const manifestPath = path.join(projectRoot, '.claude-plugin', 'plugin.json');
9
+ const conventionalHooksPath = path.join(projectRoot, 'hooks', 'hooks.json');
10
+
11
+ assert.ok(fs.existsSync(manifestPath), '.claude-plugin/plugin.json must exist');
12
+ assert.ok(fs.existsSync(conventionalHooksPath), 'hooks/hooks.json must remain available for convention-based discovery');
13
+
14
+ const manifest = JSON.parse(fs.readFileSync(manifestPath, 'utf8'));
15
+ const hooks = manifest.hooks;
16
+
17
+ if (hooks !== undefined) {
18
+ const refs = Array.isArray(hooks) ? hooks : [hooks];
19
+ const normalized = refs
20
+ .filter(ref => typeof ref === 'string')
21
+ .map(ref => ref.replace(/\\/g, '/').replace(/^\.\//, ''));
22
+
23
+ assert.ok(
24
+ !normalized.includes('hooks/hooks.json'),
25
+ 'Do not explicitly register hooks/hooks.json in .claude-plugin/plugin.json; Claude Code auto-discovers the conventional hook file and duplicate registration prevents the plugin from loading.'
26
+ );
27
+ }
28
+
29
+ console.log('PASS: Claude plugin manifest does not duplicate the conventional hooks/hooks.json registration.');