harness-everything 0.3.6 → 0.4.1
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.
- package/.agents/plugins/marketplace.json +20 -0
- package/.claude-plugin/marketplace.json +1 -1
- package/.claude-plugin/plugin.json +8 -7
- package/.github/workflows/behavioral-evals.yml +3 -3
- package/.github/workflows/ci.yml +49 -28
- package/.github/workflows/release.yml +52 -27
- package/.nvmrc +1 -0
- package/.releaserc.json +113 -0
- package/AGENTS.md +44 -24
- package/BENCHMARK_SOP.md +38 -18
- package/CHANGELOG.md +196 -3
- package/CONTRIBUTING.md +37 -9
- package/Dockerfile +1 -1
- package/PRIVACY.md +25 -0
- package/README.md +126 -91
- package/RELEASING.md +93 -0
- package/SUPPORT.md +27 -0
- package/TERMS.md +21 -0
- package/VERIFICATION.md +192 -284
- package/behavioral-evals/README.md +21 -0
- package/behavioral-evals/case-validator.js +194 -0
- package/behavioral-evals/cases/baseline-environment-detection.yaml +21 -0
- package/behavioral-evals/cases/baseline-eval-harness.yaml +27 -0
- package/behavioral-evals/cases/baseline-fable-discipline.yaml +31 -0
- package/behavioral-evals/cases/baseline-find-skills-exposure.yaml +49 -0
- package/behavioral-evals/cases/baseline-find-skills.yaml +21 -0
- package/behavioral-evals/cases/baseline-git-commit.yaml +2 -2
- package/behavioral-evals/cases/baseline-install-cognitive-os.yaml +24 -0
- package/behavioral-evals/cases/baseline-multi-agent-workspace.yaml +28 -0
- package/behavioral-evals/cases/baseline-rewrite-commits.yaml +34 -0
- package/behavioral-evals/cases/baseline-self-evolve.yaml +31 -0
- package/behavioral-evals/cases/baseline-skill-creator.yaml +20 -0
- package/behavioral-evals/cases/baseline-skill-style.yaml +21 -0
- package/behavioral-evals/cases/baseline-to-spec.yaml +34 -0
- package/behavioral-evals/cases/baseline-to-tickets.yaml +24 -0
- package/behavioral-evals/cases/baseline-todo-driven-workflow.yaml +32 -0
- package/behavioral-evals/cases/baseline-using-git-worktrees.yaml +15 -0
- package/behavioral-evals/cases/breaker-zoom-out-after-3.yaml +2 -2
- package/behavioral-evals/cases/pressure-complexity-skip.yaml +2 -2
- package/behavioral-evals/cases/pressure-scope-bypass.yaml +7 -2
- package/behavioral-evals/cases/pressure-sunk-cost-retry.yaml +7 -4
- package/behavioral-evals/cases/scope-discipline.yaml +2 -1
- package/behavioral-evals/cases/verify-before-done.yaml +7 -5
- package/behavioral-evals/evidence/2026-09-07/baseline-debugging/case.yaml +31 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-debugging/historical-case.yaml +31 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-debugging/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-debugging/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-performance/case.yaml +34 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-performance/historical-case.yaml +34 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-performance/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-performance/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-security-review/case.yaml +28 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-security-review/historical-case.yaml +28 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-security-review/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-security-review/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-simple-bugfix/case.yaml +33 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-simple-bugfix/historical-case.yaml +33 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-simple-bugfix/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/baseline-simple-bugfix/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/grill-me-adversarial/case.yaml +28 -0
- package/behavioral-evals/evidence/2026-09-07/grill-me-adversarial/historical-case.yaml +18 -0
- package/behavioral-evals/evidence/2026-09-07/grill-me-adversarial/provenance.json +30 -0
- package/behavioral-evals/evidence/2026-09-07/live-run-blocker.md +16 -0
- package/behavioral-evals/evidence/2026-09-07/manifest.json +88 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-scope-bypass/case.yaml +33 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-scope-bypass/historical-case.yaml +28 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-scope-bypass/provenance.json +30 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-docs/case.yaml +50 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-docs/historical-case.yaml +50 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-docs/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-docs/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-error-handling/case.yaml +35 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-error-handling/historical-case.yaml +35 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-error-handling/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-error-handling/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification/case.yaml +43 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification/historical-case.yaml +43 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification/result.sanitized.json +24 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-23-pressure-skip-verification/case.yaml +43 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-23-pressure-skip-verification/historical-case.yaml +43 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-23-pressure-skip-verification/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-23-pressure-skip-verification/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-27-pressure-skip-verification/case.yaml +43 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-27-pressure-skip-verification/historical-case.yaml +43 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-27-pressure-skip-verification/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-skip-verification--2026-08-27-pressure-skip-verification/result.sanitized.json +24 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry/case.yaml +50 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry/historical-case.yaml +47 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry/result.sanitized.json +24 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-23-pressure-sunk-cost-retry/case.yaml +50 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-23-pressure-sunk-cost-retry/historical-case.yaml +47 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-23-pressure-sunk-cost-retry/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-23-pressure-sunk-cost-retry/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-27-pressure-sunk-cost-retry/case.yaml +50 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-27-pressure-sunk-cost-retry/historical-case.yaml +47 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-27-pressure-sunk-cost-retry/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/pressure-sunk-cost-retry--2026-08-27-pressure-sunk-cost-retry/result.sanitized.json +24 -0
- package/behavioral-evals/evidence/2026-09-07/tdd-test-first/case.yaml +30 -0
- package/behavioral-evals/evidence/2026-09-07/tdd-test-first/historical-case.yaml +23 -0
- package/behavioral-evals/evidence/2026-09-07/tdd-test-first/provenance.json +30 -0
- package/behavioral-evals/evidence/2026-09-07/triage-matrix.json +426 -0
- package/behavioral-evals/evidence/2026-09-07/triage-matrix.md +24 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-claim-cites/case.yaml +19 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-claim-cites/historical-case.yaml +18 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-claim-cites/provenance.json +30 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done/case.yaml +33 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done/historical-case.yaml +31 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-23-verify-before-done/case.yaml +33 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-23-verify-before-done/historical-case.yaml +31 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-23-verify-before-done/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-23-verify-before-done/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-27-verify-before-done/case.yaml +33 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-27-verify-before-done/historical-case.yaml +31 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-27-verify-before-done/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-07/verify-before-done--2026-08-27-verify-before-done/result.sanitized.json +20 -0
- package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/case.yaml +33 -0
- package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/historical-case.yaml +33 -0
- package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/provenance.json +42 -0
- package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/result.sanitized.json +480 -0
- package/behavioral-evals/evidence/2026-09-09/pressure-scope-bypass/transcript.sanitized.jsonl +95 -0
- package/behavioral-evals/evidence-tool.js +443 -0
- package/behavioral-evals/fixtures/claude-denied.jsonl +3 -0
- package/behavioral-evals/fixtures/claude-legacy.json +1 -0
- package/behavioral-evals/fixtures/claude-oral-only.jsonl +2 -0
- package/behavioral-evals/fixtures/claude-success.jsonl +5 -0
- package/behavioral-evals/fixtures/claude-truncated.jsonl +3 -0
- package/behavioral-evals/fixtures/opencode-success.jsonl +6 -0
- package/behavioral-evals/results/2026-08-31-baseline-add-feature.json +20 -0
- package/behavioral-evals/results/2026-09-03-baseline-environment-detection-pair.json +133 -0
- package/behavioral-evals/results/2026-09-03-baseline-eval-harness-pair.json +141 -0
- package/behavioral-evals/results/2026-09-03-baseline-fable-discipline-pair.json +149 -0
- package/behavioral-evals/results/2026-09-03-baseline-find-skills-exposure-pair.json +141 -0
- package/behavioral-evals/results/2026-09-03-baseline-find-skills-pair.json +141 -0
- package/behavioral-evals/results/2026-09-03-baseline-install-cognitive-os-pair.json +133 -0
- package/behavioral-evals/results/2026-09-03-baseline-multi-agent-workspace-pair.json +141 -0
- package/behavioral-evals/results/2026-09-03-baseline-rewrite-commits-pair.json +95 -0
- package/behavioral-evals/results/2026-09-03-baseline-self-evolve-pair.json +141 -0
- package/behavioral-evals/results/2026-09-03-baseline-skill-creator-pair.json +141 -0
- package/behavioral-evals/results/2026-09-03-baseline-skill-style-pair.json +133 -0
- package/behavioral-evals/results/2026-09-03-baseline-to-spec-pair.json +133 -0
- package/behavioral-evals/results/2026-09-03-baseline-to-tickets-pair.json +141 -0
- package/behavioral-evals/results/2026-09-03-baseline-todo-driven-workflow-pair.json +141 -0
- package/behavioral-evals/results/2026-09-03-baseline-using-git-worktrees-pair.json +133 -0
- package/behavioral-evals/results/2026-09-03-summary-both.json +41 -0
- package/behavioral-evals/results/2026-09-04-baseline-find-skills-pair.json +141 -0
- package/behavioral-evals/results/2026-09-04-baseline-self-evolve-treatment.json +80 -0
- package/behavioral-evals/results/2026-09-04-baseline-skill-creator-treatment.json +80 -0
- package/behavioral-evals/results/2026-09-04-summary-both.json +41 -0
- package/behavioral-evals/run-with-plugin.js +15 -31
- package/behavioral-evals/run.js +446 -105
- package/behavioral-evals/transcript-parser.js +591 -0
- package/bin/cli.js +24 -12
- package/ci/ab-test-cases/grill-me-adversarial.yaml +14 -4
- package/ci/ab-test-cases/tdd-test-first.yaml +13 -6
- package/ci/ab-test-cases/verification-loop-gates.yaml +20 -7
- package/ci/ab-test-cases/verify-before-claim-cites.yaml +7 -6
- package/ci/ab-test-harness.js +113 -38
- package/ci/action-gate-platform-doc.test.js +31 -0
- package/ci/action-gate.test.js +260 -0
- package/ci/cases/case2.json +2 -2
- package/ci/consistency-check.js +51 -5
- package/ci/description-collision.js +15 -6
- package/ci/disclosure-check.js +225 -0
- package/ci/doc-capability-consistency.test.js +180 -0
- package/ci/fixtures/phase4/parallel-disjoint.stages.json +32 -0
- package/ci/fixtures/phase4/parallel-read-only.stages.json +32 -0
- package/ci/installer-roundtrip.test.js +159 -0
- package/ci/invariant-routing.test.js +74 -0
- package/ci/issue99-runtime.test.js +218 -0
- package/ci/mechanism-2aa-codex-action-gate-policy.test.js +117 -0
- package/ci/mechanism-2ab-openai-plugin-sync-integrity.test.js +80 -0
- package/ci/mechanism-2j-behavioral-runner.test.js +43 -1
- package/ci/mechanism-2k-reference-check.test.js +57 -2
- package/ci/mechanism-2l-multi-agent-workspace.test.js +202 -6
- package/ci/mechanism-2m-self-repo-guard.test.js +65 -0
- package/ci/mechanism-2n-opencode-plugin.test.js +461 -0
- package/ci/mechanism-2o-state-home-isolation.test.js +183 -0
- package/ci/mechanism-2p-installer-links.test.js +183 -0
- package/ci/mechanism-2q-installer-global-shared-store.test.js +132 -0
- package/ci/mechanism-2r-uninstall-flags.test.js +148 -0
- package/ci/mechanism-2s-tdd-quality.test.js +203 -0
- package/ci/mechanism-2t-disclosure.test.js +90 -0
- package/ci/mechanism-2t-evidence-triage.test.js +200 -0
- package/ci/mechanism-2t-issue-20.test.js +75 -0
- package/ci/mechanism-2t-platform-skill-targets.test.js +164 -0
- package/ci/mechanism-2t-transcript-parser.test.js +179 -0
- package/ci/mechanism-2t-workspace-identity.test.js +200 -0
- package/ci/mechanism-2u-issue-42-regressions.test.js +239 -0
- package/ci/mechanism-2u-issue-44.test.js +35 -0
- package/ci/mechanism-2v-verifier-boundaries.test.js +138 -0
- package/ci/mechanism-2w-platform-compatibility.test.js +148 -0
- package/ci/mechanism-2x-release-automation.test.js +157 -0
- package/ci/mechanism-2x-release-commit-message.test.js +66 -0
- package/ci/mechanism-2y-release-tooling-compat.test.js +47 -0
- package/ci/mechanism-2z-claude-plugin-manifest.test.js +29 -0
- package/ci/mutation-check.js +53 -0
- package/ci/mutation-loader.js +57 -0
- package/ci/negative-controls.js +76 -46
- package/ci/openai-plugin-package.test.js +182 -0
- package/ci/openai-plugin-runtime.test.js +175 -0
- package/ci/openai-plugin-submission.test.js +101 -0
- package/ci/phase4-ensemble-memory.test.js +307 -0
- package/ci/reference-check.js +73 -25
- package/ci/release-consistency-check.js +24 -1
- package/ci/repository-contract.test.js +17 -0
- package/ci/router-phase2-safety.test.js +81 -0
- package/ci/router-workflow-plan.test.js +264 -0
- package/ci/routing-check.js +6 -5
- package/ci/runner.js +6 -5
- package/ci/skill-routing-check.js +48 -11
- package/ci/test-behavioral-case-validation.js +62 -0
- package/ci/test-helper.js +10 -0
- package/ci/workflow-plan-consumers-edge.test.js +72 -0
- package/ci/workflow-plan-consumers.test.js +276 -0
- package/docs/architecture.md +137 -57
- package/docs/audit.md +22 -20
- package/docs/codex-action-gate.md +90 -0
- package/docs/issue-20-rollup.md +47 -0
- package/docs/mechanism-first-skill-mesh.md +22 -20
- package/docs/openai-plugin.md +118 -0
- package/docs/philosophy.md +28 -17
- package/docs/platform-capabilities.md +73 -0
- package/docs/platform-compatibility.json +270 -0
- package/docs/reflection.md +37 -22
- package/docs/release-evidence.md +27 -0
- package/docs/repository-contract.md +27 -0
- package/docs/routing.md +157 -44
- package/docs/security-audit.md +38 -0
- package/docs/troubleshooting.md +94 -51
- package/docs/verifier-boundaries.md +45 -0
- package/docs/workflows/fable-mode.md +77 -21
- package/docs/workflows/find-skills.md +40 -0
- package/docs/workflows/grill-with-docs.md +2 -2
- package/docs/workflows/multi-agent-workspace.md +22 -10
- package/docs/workflows/self-evolve.md +33 -0
- package/docs/workflows/tdd.md +36 -0
- package/docs/workflows/todo-driven-workflow.md +33 -0
- package/environment-detection/SKILL.md +3 -3
- package/eval-harness/SKILL.md +7 -7
- package/eval-harness/scripts/evaluate.js +12 -10
- package/evals/evaluation-reports.md +14 -0
- package/evals/harness-everything/tasks/positive-tier1.yaml +1 -1
- package/evals/harness-everything/tasks/positive-tier2.yaml +1 -1
- package/evals/harness-everything/tasks/positive-tier3.yaml +1 -1
- package/evals/install-cognitive-os/tasks/positive-trigger.yaml +1 -1
- package/fable-discipline/SKILL.md +2 -2
- package/fable-mode/CONTRACT-FORMAT.md +147 -33
- package/fable-mode/SKILL.md +3 -3
- package/fable-mode/agents/fable-orchestrator.md +95 -35
- package/fable-mode/execution-guardrails/SKILL.md +2 -2
- package/fable-mode/fable-haiku/SKILL.md +3 -3
- package/fable-mode/fable-opus/SKILL.md +3 -3
- package/fable-mode/fable-sonnet/SKILL.md +3 -3
- package/fable-mode/scripts/ensemble-review.js +371 -0
- package/fable-mode/scripts/model-selector.js +134 -3
- package/fable-mode/scripts/workflow-plan-consumer.js +337 -0
- package/find-skills/SKILL.md +5 -5
- package/find-skills/references/discovery-flow.md +1 -1
- package/find-skills/scripts/use-skill.js +7 -1
- package/git-commit/SKILL.md +4 -11
- package/grill-me/SKILL.md +5 -5
- package/grill-with-docs/SKILL.md +7 -7
- package/grill-with-docs/references/session-playbook.md +6 -4
- package/harness-everything/SKILL.md +23 -22
- package/harness-everything/references/router-workflow-plan.md +128 -0
- package/harness-everything/references/triage-and-tiers.md +173 -96
- package/harness-everything/schemas/router-task-shape.schema.json +126 -0
- package/harness-everything/schemas/router-workflow-plan.schema.json +149 -0
- package/harness-everything/scripts/bootstrap.js +13 -4
- package/harness-everything/scripts/ensemble-policy.js +102 -0
- package/harness-everything/scripts/kernel-router.js +212 -0
- package/harness-everything/scripts/router-contract.js +612 -0
- package/harness-everything/scripts/self-heal.js +29 -20
- package/harness-everything/scripts/tier-router.js +346 -221
- package/hooks/hooks.json +60 -15
- package/hooks/scripts/action-gate-rules.json +77 -0
- package/hooks/scripts/action-gate.js +528 -0
- package/hooks/scripts/atomic-commit-check.js +5 -5
- package/hooks/scripts/codex-action-gate-post.js +46 -0
- package/hooks/scripts/codex-permission-request.js +140 -0
- package/hooks/scripts/context-compact.js +4 -3
- package/hooks/scripts/contract-test.js +118 -51
- package/hooks/scripts/lib/fable-contracts.js +121 -0
- package/hooks/scripts/lib/harness-state.js +164 -28
- package/hooks/scripts/lib/platforms/claude.js +45 -62
- package/hooks/scripts/lib/platforms/codex.js +35 -61
- package/hooks/scripts/lib/platforms/continue.js +7 -7
- package/hooks/scripts/lib/platforms/copilot.js +29 -53
- package/hooks/scripts/lib/platforms/hermes.js +71 -20
- package/hooks/scripts/openai-session-start.js +81 -0
- package/hooks/scripts/rule-of-3-tracker.js +9 -8
- package/hooks/scripts/rule-of-3.js +29 -21
- package/hooks/scripts/state-persist.js +5 -4
- package/hooks/scripts/stop-gate.js +2 -2
- package/hooks/scripts/subagent-scope-guard.js +144 -38
- package/hooks/scripts/subagent-start.js +19 -0
- package/hooks/scripts/subagent-stop.js +6 -0
- package/improve-codebase-architecture/SKILL.md +2 -4
- package/improve-codebase-architecture/references/README.md +16 -0
- package/install-cognitive-os/SKILL.md +28 -22
- package/multi-agent-workspace/SKILL.md +16 -15
- package/multi-agent-workspace/references/architecture-guide.md +13 -9
- package/multi-agent-workspace/references/orchestration.md +39 -22
- package/multi-agent-workspace/scripts/consume-workflow-plan.js +129 -0
- package/multi-agent-workspace/scripts/index_memory.js +79 -26
- package/multi-agent-workspace/scripts/path-boundary.js +59 -0
- package/multi-agent-workspace/scripts/project-docs-resolver.js +357 -0
- package/multi-agent-workspace/scripts/scaffold.js +258 -103
- package/multi-agent-workspace/templates/AGENTS.md +20 -0
- package/multi-agent-workspace/workflows/03-generation.md +8 -10
- package/multi-agent-workspace/workflows/04-launcher.md +5 -5
- package/opencode-plugin/README.md +114 -71
- package/opencode-plugin/hooks/verify.js +62 -12
- package/opencode-plugin/index.mjs +547 -0
- package/opencode-plugin/plugin.json +4 -12
- package/package.json +20 -4
- package/plugins/harness-everything/.codex-plugin/plugin.json +39 -0
- package/plugins/harness-everything/hooks/hooks.json +197 -0
- package/plugins/harness-everything/hooks/scripts/action-gate-rules.json +77 -0
- package/plugins/harness-everything/hooks/scripts/action-gate.js +528 -0
- package/plugins/harness-everything/hooks/scripts/atomic-commit-check.js +68 -0
- package/plugins/harness-everything/hooks/scripts/boundary-guard.js +55 -0
- package/plugins/harness-everything/hooks/scripts/codex-action-gate-post.js +46 -0
- package/plugins/harness-everything/hooks/scripts/codex-permission-request.js +140 -0
- package/plugins/harness-everything/hooks/scripts/context-compact.js +115 -0
- package/plugins/harness-everything/hooks/scripts/contract-test.js +146 -0
- package/plugins/harness-everything/hooks/scripts/depth-guard.js +65 -0
- package/plugins/harness-everything/hooks/scripts/lib/fable-contracts.js +121 -0
- package/plugins/harness-everything/hooks/scripts/lib/harness-state.js +275 -0
- package/plugins/harness-everything/hooks/scripts/lib/platforms/claude.js +168 -0
- package/plugins/harness-everything/hooks/scripts/lib/platforms/codex.js +88 -0
- package/plugins/harness-everything/hooks/scripts/lib/platforms/continue.js +91 -0
- package/plugins/harness-everything/hooks/scripts/lib/platforms/copilot.js +85 -0
- package/plugins/harness-everything/hooks/scripts/lib/platforms/cursor.js +89 -0
- package/plugins/harness-everything/hooks/scripts/lib/platforms/hermes.js +108 -0
- package/plugins/harness-everything/hooks/scripts/lib/platforms/index.js +17 -0
- package/plugins/harness-everything/hooks/scripts/lib/platforms/worktrees.js +20 -0
- package/plugins/harness-everything/hooks/scripts/lib/workspace.js +290 -0
- package/plugins/harness-everything/hooks/scripts/openai-session-start.js +81 -0
- package/plugins/harness-everything/hooks/scripts/reset-circuit-breaker.js +30 -0
- package/plugins/harness-everything/hooks/scripts/rule-of-3-tracker.js +125 -0
- package/plugins/harness-everything/hooks/scripts/rule-of-3.js +195 -0
- package/plugins/harness-everything/hooks/scripts/state-persist.js +100 -0
- package/plugins/harness-everything/hooks/scripts/stop-gate.js +80 -0
- package/plugins/harness-everything/hooks/scripts/subagent-scope-guard.js +194 -0
- package/plugins/harness-everything/hooks/scripts/subagent-start.js +19 -0
- package/plugins/harness-everything/hooks/scripts/subagent-stop.js +6 -0
- package/plugins/harness-everything/hooks/session-start.js +81 -0
- package/plugins/harness-everything/plugin.json +43 -0
- package/plugins/harness-everything/scripts/lib/advisory-text.js +174 -0
- package/plugins/harness-everything/skills/environment-detection/SKILL.md +41 -0
- package/plugins/harness-everything/skills/environment-detection/references/shell-syntax-rules.md +76 -0
- package/plugins/harness-everything/skills/environment-detection/scripts/preflight.js +103 -0
- package/plugins/harness-everything/skills/eval-harness/SKILL.md +42 -0
- package/plugins/harness-everything/skills/eval-harness/references/scoring-rubric.md +68 -0
- package/plugins/harness-everything/skills/eval-harness/scripts/evaluate.js +79 -0
- package/plugins/harness-everything/skills/fable-discipline/SKILL.md +41 -0
- package/plugins/harness-everything/skills/fable-discipline/references/discipline-rules.md +22 -0
- package/plugins/harness-everything/skills/fable-mode/CONTRACT-FORMAT.md +165 -0
- package/plugins/harness-everything/skills/fable-mode/SKILL.md +39 -0
- package/plugins/harness-everything/skills/fable-mode/agents/fable-orchestrator.md +159 -0
- package/plugins/harness-everything/skills/fable-mode/agents/fable-verifier.md +32 -0
- package/plugins/harness-everything/skills/fable-mode/agents/fable-worker-haiku.md +52 -0
- package/plugins/harness-everything/skills/fable-mode/agents/fable-worker-sonnet.md +53 -0
- package/plugins/harness-everything/skills/fable-mode/execution-guardrails/SKILL.md +43 -0
- package/plugins/harness-everything/skills/fable-mode/execution-guardrails/references/guardrail-rules.md +56 -0
- package/plugins/harness-everything/skills/fable-mode/fable-haiku/SKILL.md +36 -0
- package/plugins/harness-everything/skills/fable-mode/fable-haiku/references/routing-notes.md +26 -0
- package/plugins/harness-everything/skills/fable-mode/fable-opus/SKILL.md +36 -0
- package/plugins/harness-everything/skills/fable-mode/fable-sonnet/SKILL.md +36 -0
- package/plugins/harness-everything/skills/fable-mode/model-matrix.json +32 -0
- package/plugins/harness-everything/skills/fable-mode/references/execution-phases.md +37 -0
- package/plugins/harness-everything/skills/fable-mode/references/model-matrix.md +39 -0
- package/plugins/harness-everything/skills/fable-mode/scripts/ensemble-review.js +371 -0
- package/plugins/harness-everything/skills/fable-mode/scripts/model-selector.js +260 -0
- package/plugins/harness-everything/skills/fable-mode/scripts/workflow-plan-consumer.js +337 -0
- package/plugins/harness-everything/skills/find-skills/SKILL.md +48 -0
- package/plugins/harness-everything/skills/find-skills/references/discovery-flow.md +102 -0
- package/plugins/harness-everything/skills/find-skills/scripts/use-skill.js +112 -0
- package/plugins/harness-everything/skills/git-commit/SKILL.md +38 -0
- package/plugins/harness-everything/skills/git-commit/guides/ANGULAR_STYLE.md +61 -0
- package/plugins/harness-everything/skills/git-commit/guides/COMMIT_GENERATION.md +24 -0
- package/plugins/harness-everything/skills/git-commit/guides/LANGUAGE_DETECTION.md +16 -0
- package/plugins/harness-everything/skills/git-commit/guides/MAIN_REPO.md +15 -0
- package/plugins/harness-everything/skills/git-commit/guides/SUBMODULES.md +27 -0
- package/plugins/harness-everything/skills/git-commit/references/commit-flow.md +30 -0
- package/plugins/harness-everything/skills/grill-me/SKILL.md +43 -0
- package/plugins/harness-everything/skills/grill-me/references/grilling-playbook.md +38 -0
- package/plugins/harness-everything/skills/grill-with-docs/ADR-FORMAT.md +47 -0
- package/plugins/harness-everything/skills/grill-with-docs/CONTEXT-FORMAT.md +63 -0
- package/plugins/harness-everything/skills/grill-with-docs/SKILL.md +41 -0
- package/plugins/harness-everything/skills/grill-with-docs/references/session-playbook.md +99 -0
- package/plugins/harness-everything/skills/harness-everything/SKILL.md +43 -0
- package/plugins/harness-everything/skills/harness-everything/examples/README.md +9 -0
- package/plugins/harness-everything/skills/harness-everything/examples/todo-workflow.md +22 -0
- package/plugins/harness-everything/skills/harness-everything/references/router-workflow-plan.md +128 -0
- package/plugins/harness-everything/skills/harness-everything/references/skill-registry.md +39 -0
- package/plugins/harness-everything/skills/harness-everything/references/triage-and-tiers.md +193 -0
- package/plugins/harness-everything/skills/harness-everything/schemas/router-task-shape.schema.json +126 -0
- package/plugins/harness-everything/skills/harness-everything/schemas/router-workflow-plan.schema.json +149 -0
- package/plugins/harness-everything/skills/harness-everything/scripts/bootstrap.js +212 -0
- package/plugins/harness-everything/skills/harness-everything/scripts/ensemble-policy.js +102 -0
- package/plugins/harness-everything/skills/harness-everything/scripts/kernel-router.js +212 -0
- package/plugins/harness-everything/skills/harness-everything/scripts/router-contract.js +612 -0
- package/plugins/harness-everything/skills/harness-everything/scripts/routing-keywords.json +320 -0
- package/plugins/harness-everything/skills/harness-everything/scripts/self-heal.js +250 -0
- package/plugins/harness-everything/skills/harness-everything/scripts/tier-router.js +459 -0
- package/plugins/harness-everything/skills/harness-everything/scripts/verify-gate.js +100 -0
- package/plugins/harness-everything/skills/improve-codebase-architecture/SKILL.md +38 -0
- package/plugins/harness-everything/skills/improve-codebase-architecture/guides/DEEPENING.md +37 -0
- package/plugins/harness-everything/skills/improve-codebase-architecture/guides/HTML-REPORT.md +123 -0
- package/plugins/harness-everything/skills/improve-codebase-architecture/guides/INTERFACE-DESIGN.md +44 -0
- package/plugins/harness-everything/skills/improve-codebase-architecture/guides/LANGUAGE.md +53 -0
- package/plugins/harness-everything/skills/improve-codebase-architecture/guides/refactoring-workflow.md +59 -0
- package/plugins/harness-everything/skills/improve-codebase-architecture/references/README.md +16 -0
- package/plugins/harness-everything/skills/install-cognitive-os/SKILL.md +44 -0
- package/plugins/harness-everything/skills/install-cognitive-os/references/cognitive-loop.md +27 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/NOTICE-agency-agents.md +27 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/SKILL.md +38 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/examples/csharp-monorepo.md +123 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/examples/typescript-cli.md +126 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-claude.md +77 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-codex.md +146 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-continue.md +38 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-copilot.md +74 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-cursor.md +40 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/platform-hermes.md +38 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/guidelines/role-capabilities.md +108 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/references/agency-agents.md +31 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/references/architecture-guide.md +36 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/references/orchestration.md +65 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/scripts/agency-catalog.js +190 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/scripts/consume-workflow-plan.js +129 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/scripts/index_memory.js +104 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/scripts/path-boundary.js +59 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/scripts/project-docs-resolver.js +357 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/scripts/scaffold.js +408 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/AGENTS.md +20 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/README.md +17 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/agent-launcher.md +68 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/backend-developer-csharp.md +56 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/backend-developer-polyglot.md +62 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/backend-developer-ts.md +57 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/memory-keeper.md +37 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/requirement-analyzer.md +66 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/specialist-agent.md +34 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/task-verifier.md +56 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/templates/tool-maker.md +33 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/workflows/01-init.md +22 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/workflows/02-analysis.md +23 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/workflows/03-generation.md +25 -0
- package/plugins/harness-everything/skills/multi-agent-workspace/workflows/04-launcher.md +16 -0
- package/plugins/harness-everything/skills/repo-docs/SKILL.md +42 -0
- package/plugins/harness-everything/skills/repo-docs/references/process-guide.md +60 -0
- package/plugins/harness-everything/skills/repo-docs/templates/agents-template.md +264 -0
- package/plugins/harness-everything/skills/repo-docs/templates/knowledge-agents-template.md +50 -0
- package/plugins/harness-everything/skills/repo-docs/templates/knowledge-readme-template.md +83 -0
- package/plugins/harness-everything/skills/repo-docs/templates/manual-recon.md +117 -0
- package/plugins/harness-everything/skills/repo-docs/templates/multi-skills-readme-template.md +72 -0
- package/plugins/harness-everything/skills/repo-docs/templates/product-agents-template.md +78 -0
- package/plugins/harness-everything/skills/repo-docs/templates/product-readme-template.md +95 -0
- package/plugins/harness-everything/skills/repo-docs/templates/readme-template.md +136 -0
- package/plugins/harness-everything/skills/repo-docs/templates/single-skill-readme-template.md +70 -0
- package/plugins/harness-everything/skills/repo-docs/templates/skills-agents-template.md +60 -0
- package/plugins/harness-everything/skills/rewrite-commits/SKILL.md +37 -0
- package/plugins/harness-everything/skills/rewrite-commits/references/workflow.md +18 -0
- package/plugins/harness-everything/skills/security-review/SKILL.md +47 -0
- package/plugins/harness-everything/skills/security-review/guides/OWASP-PATTERNS.md +31 -0
- package/plugins/harness-everything/skills/security-review/guides/STRIDE-THREAT-MODEL.md +21 -0
- package/plugins/harness-everything/skills/security-review/references/security-checklist.md +478 -0
- package/plugins/harness-everything/skills/security-review/scripts/audit-secrets.js +85 -0
- package/plugins/harness-everything/skills/self-evolve/SKILL.md +45 -0
- package/plugins/harness-everything/skills/self-evolve/references/memory-resolution.md +61 -0
- package/plugins/harness-everything/skills/self-evolve/scripts/persist-memory.js +272 -0
- package/plugins/harness-everything/skills/self-evolve/scripts/register-dynamic-skill.js +253 -0
- package/plugins/harness-everything/skills/self-evolve/scripts/self-regression.js +125 -0
- package/plugins/harness-everything/skills/skill-creator/SKILL.md +47 -0
- package/plugins/harness-everything/skills/skill-creator/references/authoring-workflow.md +14 -0
- package/plugins/harness-everything/skills/skill-creator/references/dynamic-generation-contract.md +21 -0
- package/plugins/harness-everything/skills/skill-creator/references/quality-checklist.md +16 -0
- package/plugins/harness-everything/skills/skill-creator/references/quality-principles.md +81 -0
- package/plugins/harness-everything/skills/skill-creator/references/testing-workflow.md +27 -0
- package/plugins/harness-everything/skills/skill-style/SKILL.md +42 -0
- package/plugins/harness-everything/skills/skill-style/references/style-guide.md +62 -0
- package/plugins/harness-everything/skills/tdd/SKILL.md +38 -0
- package/plugins/harness-everything/skills/tdd/fixtures/pass-evidence.json +100 -0
- package/plugins/harness-everything/skills/tdd/fixtures/quality-scenarios.json +20 -0
- package/plugins/harness-everything/skills/tdd/guides/deep-modules.md +33 -0
- package/plugins/harness-everything/skills/tdd/guides/interface-design.md +31 -0
- package/plugins/harness-everything/skills/tdd/guides/mocking.md +59 -0
- package/plugins/harness-everything/skills/tdd/guides/refactoring.md +10 -0
- package/plugins/harness-everything/skills/tdd/guides/tests.md +61 -0
- package/plugins/harness-everything/skills/tdd/references/core-discipline.md +51 -0
- package/plugins/harness-everything/skills/tdd/references/integration-testing.md +13 -0
- package/plugins/harness-everything/skills/tdd/references/quality-model.md +68 -0
- package/plugins/harness-everything/skills/tdd/references/unit-testing.md +15 -0
- package/plugins/harness-everything/skills/tdd/schemas/quality-evidence.schema.json +211 -0
- package/plugins/harness-everything/skills/tdd/schemas/quality-report.schema.json +110 -0
- package/plugins/harness-everything/skills/tdd/scripts/quality-gate.js +475 -0
- package/plugins/harness-everything/skills/to-spec/SKILL.md +40 -0
- package/plugins/harness-everything/skills/to-spec/references/process.md +77 -0
- package/plugins/harness-everything/skills/to-spec/scripts/check-project-docs.js +239 -0
- package/plugins/harness-everything/skills/to-spec/scripts/project-docs-resolver.js +111 -0
- package/plugins/harness-everything/skills/to-spec/templates/cli-reference.md +31 -0
- package/plugins/harness-everything/skills/to-spec/templates/dev-doc.md +27 -0
- package/plugins/harness-everything/skills/to-spec/templates/feature-spec.md +59 -0
- package/plugins/harness-everything/skills/to-spec/templates/schema-doc.md +31 -0
- package/plugins/harness-everything/skills/to-tickets/SKILL.md +42 -0
- package/plugins/harness-everything/skills/to-tickets/references/publishing-and-templates.md +97 -0
- package/plugins/harness-everything/skills/todo-driven-workflow/SKILL.md +42 -0
- package/plugins/harness-everything/skills/todo-driven-workflow/references/execution-guide.md +27 -0
- package/plugins/harness-everything/skills/using-git-worktrees/SKILL.md +49 -0
- package/plugins/harness-everything/skills/using-git-worktrees/references/workflow-details.md +108 -0
- package/plugins/harness-everything/skills/verification-loop/SKILL.md +46 -0
- package/plugins/harness-everything/skills/verification-loop/references/verification-phases.md +116 -0
- package/plugins/harness-everything/skills/verification-loop/templates/verification-report.template.md +15 -0
- package/plugins/harness-everything/skills/verify-before-claim/SKILL.md +43 -0
- package/plugins/harness-everything/skills/verify-before-claim/references/verification-guide.md +73 -0
- package/plugins/harness-everything/skills/zoom-out/SKILL.md +42 -0
- package/plugins/harness-everything/skills/zoom-out/references/circuit-breaker.md +66 -0
- package/plugins/harness-everything/skills/zoom-out/templates/zoom-out-report.template.md +20 -0
- package/repo-docs/SKILL.md +3 -7
- package/rewrite-commits/SKILL.md +2 -2
- package/scripts/build-openai-submission.js +158 -0
- package/scripts/installer.js +86 -11
- package/scripts/lib/advisory-text.js +25 -18
- package/scripts/lib/execution-contract.js +70 -0
- package/scripts/lib/manifest.js +34 -38
- package/scripts/lib/path-boundary.js +90 -0
- package/scripts/lib/skills.js +124 -70
- package/scripts/lib/workspace.js +269 -7
- package/scripts/repository-contract.js +213 -0
- package/scripts/sync-current-docs.js +12 -0
- package/scripts/sync-openai-plugin.js +91 -0
- package/scripts/sync-release-version.js +212 -0
- package/security-review/SKILL.md +4 -7
- package/security-review/scripts/audit-secrets.js +9 -8
- package/self-evolve/SKILL.md +9 -11
- package/self-evolve/references/memory-resolution.md +4 -1
- package/self-evolve/scripts/persist-memory.js +221 -88
- package/self-evolve/scripts/register-dynamic-skill.js +12 -10
- package/self-evolve/scripts/self-regression.js +23 -8
- package/skill-creator/SKILL.md +12 -14
- package/skill-creator/references/quality-checklist.md +1 -0
- package/skill-style/SKILL.md +4 -4
- package/skill-style/references/style-guide.md +29 -3
- package/submission/openai/README.md +67 -0
- package/submission/openai/listing.json +31 -0
- package/submission/openai/test-cases.json +61 -0
- package/tdd/SKILL.md +20 -26
- package/tdd/fixtures/pass-evidence.json +100 -0
- package/tdd/fixtures/quality-scenarios.json +20 -0
- package/tdd/references/core-discipline.md +2 -0
- package/tdd/references/integration-testing.md +13 -0
- package/tdd/references/quality-model.md +68 -0
- package/tdd/references/unit-testing.md +15 -0
- package/tdd/schemas/quality-evidence.schema.json +211 -0
- package/tdd/schemas/quality-report.schema.json +110 -0
- package/tdd/scripts/quality-gate.js +475 -0
- package/to-spec/SKILL.md +3 -5
- package/to-spec/scripts/check-project-docs.js +42 -62
- package/to-spec/scripts/project-docs-resolver.js +111 -0
- package/to-tickets/SKILL.md +4 -4
- package/todo-driven-workflow/SKILL.md +6 -4
- package/using-git-worktrees/SKILL.md +3 -3
- package/verification-loop/SKILL.md +4 -4
- package/verify-before-claim/SKILL.md +2 -2
- package/zoom-out/SKILL.md +4 -4
- package/ab-test-deep-analysis-2026-08-27.md +0 -383
- package/ab-test-results-2026-08-27.md +0 -152
- package/ab-test-statistical-analysis-31cases.md +0 -193
- package/opencode-plugin/hooks/circuit-breaker.js +0 -121
- package/opencode-plugin/hooks/compliance.js +0 -128
- package/opencode-plugin/hooks/post-edit.js +0 -62
- package/opencode-plugin/hooks/pre-complete.js +0 -65
- package/opencode-plugin/test-hook-firing.js +0 -198
|
@@ -0,0 +1,475 @@
|
|
|
1
|
+
#!/usr/bin/env node
|
|
2
|
+
|
|
3
|
+
const fs = require('fs');
|
|
4
|
+
const path = require('path');
|
|
5
|
+
|
|
6
|
+
const VERSION = '1.0.0';
|
|
7
|
+
const DIMENSIONS = [
|
|
8
|
+
'positiveParameters',
|
|
9
|
+
'negativeParameters',
|
|
10
|
+
'inputCompleteness',
|
|
11
|
+
'outputCompleteness',
|
|
12
|
+
'errorHandling',
|
|
13
|
+
'unexpectedInput',
|
|
14
|
+
'determinism',
|
|
15
|
+
'sourceTraceability'
|
|
16
|
+
];
|
|
17
|
+
const WEIGHTS = {
|
|
18
|
+
positiveParameters: 0.15,
|
|
19
|
+
negativeParameters: 0.15,
|
|
20
|
+
inputCompleteness: 0.15,
|
|
21
|
+
outputCompleteness: 0.15,
|
|
22
|
+
errorHandling: 0.15,
|
|
23
|
+
unexpectedInput: 0.10,
|
|
24
|
+
determinism: 0.10,
|
|
25
|
+
sourceTraceability: 0.05
|
|
26
|
+
};
|
|
27
|
+
const REQUIRED_CLASSES = {
|
|
28
|
+
inputCompleteness: ['required', 'optional', 'omitted', 'empty', 'null', 'min', 'max', 'boundary'],
|
|
29
|
+
outputCompleteness: ['value', 'type', 'schema', 'required-field', 'prohibited-extra-field', 'side-effect'],
|
|
30
|
+
errorHandling: ['error-type', 'error-code', 'message', 'state-cleanup', 'rollback', 'no-unintended-output'],
|
|
31
|
+
unexpectedInput: ['malformed', 'unknown', 'extra', 'unsupported', 'out-of-range', 'duplicate', 'wrong-type']
|
|
32
|
+
};
|
|
33
|
+
const CHECK_STATUSES = new Set(['PASS', 'FAIL', 'SKIPPED', 'N/A']);
|
|
34
|
+
const SIDE_EFFECT_FIELDS = ['createdFiles', 'changedFiles', 'deletedFiles', 'persistentState', 'events', 'logs', 'externalCalls'];
|
|
35
|
+
|
|
36
|
+
function isObject(value) {
|
|
37
|
+
return value !== null && typeof value === 'object' && !Array.isArray(value);
|
|
38
|
+
}
|
|
39
|
+
|
|
40
|
+
function nonEmpty(value) {
|
|
41
|
+
return typeof value === 'string' && value.trim().length > 0;
|
|
42
|
+
}
|
|
43
|
+
|
|
44
|
+
function round(value) {
|
|
45
|
+
return Math.round((value + Number.EPSILON) * 100) / 100;
|
|
46
|
+
}
|
|
47
|
+
|
|
48
|
+
function stable(value) {
|
|
49
|
+
if (Array.isArray(value)) return value.map(stable);
|
|
50
|
+
if (!isObject(value)) return value;
|
|
51
|
+
return Object.keys(value).sort().reduce((result, key) => {
|
|
52
|
+
result[key] = stable(value[key]);
|
|
53
|
+
return result;
|
|
54
|
+
}, {});
|
|
55
|
+
}
|
|
56
|
+
|
|
57
|
+
function equal(left, right) {
|
|
58
|
+
return JSON.stringify(stable(left)) === JSON.stringify(stable(right));
|
|
59
|
+
}
|
|
60
|
+
|
|
61
|
+
function clone(value) {
|
|
62
|
+
return value === undefined ? undefined : JSON.parse(JSON.stringify(value));
|
|
63
|
+
}
|
|
64
|
+
|
|
65
|
+
function normalizeRun(run, normalizations, errors, prefix) {
|
|
66
|
+
const comparable = {
|
|
67
|
+
exitStatus: run.exitStatus,
|
|
68
|
+
errorCategory: run.errorCategory,
|
|
69
|
+
output: clone(run.output),
|
|
70
|
+
sideEffects: clone(run.sideEffects)
|
|
71
|
+
};
|
|
72
|
+
|
|
73
|
+
for (const normalization of normalizations) {
|
|
74
|
+
if (!isObject(normalization) || !nonEmpty(normalization.path) || !nonEmpty(normalization.rule) || !nonEmpty(normalization.sourceSection)) {
|
|
75
|
+
errors.push(`${prefix}.normalizations entries require path, rule, and sourceSection`);
|
|
76
|
+
continue;
|
|
77
|
+
}
|
|
78
|
+
const parts = normalization.path.split('.');
|
|
79
|
+
if (!['output', 'sideEffects'].includes(parts[0]) || parts.some(part => !part)) {
|
|
80
|
+
errors.push(`${prefix}.normalizations path must start with output or sideEffects`);
|
|
81
|
+
continue;
|
|
82
|
+
}
|
|
83
|
+
let cursor = comparable;
|
|
84
|
+
let found = true;
|
|
85
|
+
for (let index = 0; index < parts.length - 1; index++) {
|
|
86
|
+
if (!isObject(cursor) && !Array.isArray(cursor)) { found = false; break; }
|
|
87
|
+
if (!Object.prototype.hasOwnProperty.call(cursor, parts[index])) { found = false; break; }
|
|
88
|
+
cursor = cursor[parts[index]];
|
|
89
|
+
}
|
|
90
|
+
const leaf = parts[parts.length - 1];
|
|
91
|
+
if (!found || (!isObject(cursor) && !Array.isArray(cursor)) || !Object.prototype.hasOwnProperty.call(cursor, leaf)) {
|
|
92
|
+
errors.push(`${prefix}.normalizations path does not exist in every run: ${normalization.path}`);
|
|
93
|
+
continue;
|
|
94
|
+
}
|
|
95
|
+
cursor[leaf] = `[normalized: ${normalization.rule}]`;
|
|
96
|
+
}
|
|
97
|
+
return comparable;
|
|
98
|
+
}
|
|
99
|
+
|
|
100
|
+
function emptyDimension(applicable = false) {
|
|
101
|
+
return { applicable, passed: 0, required: 0, percentage: applicable ? 0 : null };
|
|
102
|
+
}
|
|
103
|
+
|
|
104
|
+
function evaluateChecks(name, dimension, errors, prefix, skippedTests, naDeclarations) {
|
|
105
|
+
if (!isObject(dimension) || typeof dimension.applicable !== 'boolean') {
|
|
106
|
+
errors.push(`${prefix}.${name} requires an explicit boolean applicable field`);
|
|
107
|
+
return emptyDimension(true);
|
|
108
|
+
}
|
|
109
|
+
if (!dimension.applicable) {
|
|
110
|
+
if (!nonEmpty(dimension.reason)) errors.push(`${prefix}.${name} N/A requires a written reason`);
|
|
111
|
+
const declaration = { dimension: name, reason: dimension.reason || '' };
|
|
112
|
+
naDeclarations.push(declaration);
|
|
113
|
+
return { ...emptyDimension(false), reason: dimension.reason || '' };
|
|
114
|
+
}
|
|
115
|
+
if (!Array.isArray(dimension.checks) || dimension.checks.length === 0) {
|
|
116
|
+
errors.push(`${prefix}.${name} requires at least one executed check`);
|
|
117
|
+
return { ...emptyDimension(true), checks: [] };
|
|
118
|
+
}
|
|
119
|
+
|
|
120
|
+
const seenIds = new Set();
|
|
121
|
+
const seenClasses = new Set();
|
|
122
|
+
let passed = 0;
|
|
123
|
+
let required = 0;
|
|
124
|
+
let executed = 0;
|
|
125
|
+
const checks = dimension.checks.map((check, index) => {
|
|
126
|
+
const location = `${prefix}.${name}.checks[${index}]`;
|
|
127
|
+
if (!isObject(check)) {
|
|
128
|
+
errors.push(`${location} must be an object`);
|
|
129
|
+
return check;
|
|
130
|
+
}
|
|
131
|
+
if (!nonEmpty(check.id)) errors.push(`${location}.id is required`);
|
|
132
|
+
else if (seenIds.has(check.id)) errors.push(`${location}.id is duplicated: ${check.id}`);
|
|
133
|
+
else seenIds.add(check.id);
|
|
134
|
+
if (!nonEmpty(check.class)) errors.push(`${location}.class is required`);
|
|
135
|
+
else seenClasses.add(check.class);
|
|
136
|
+
if (!CHECK_STATUSES.has(check.status)) errors.push(`${location}.status must be PASS, FAIL, SKIPPED, or N/A`);
|
|
137
|
+
if (check.status === 'N/A') {
|
|
138
|
+
if (!nonEmpty(check.reason)) errors.push(`${location} N/A requires a written reason`);
|
|
139
|
+
naDeclarations.push({ dimension: name, checkId: check.id || `index-${index}`, reason: check.reason || '' });
|
|
140
|
+
} else {
|
|
141
|
+
required++;
|
|
142
|
+
if (check.status === 'PASS') passed++;
|
|
143
|
+
if (check.status === 'PASS' || check.status === 'FAIL') executed++;
|
|
144
|
+
if (check.status === 'SKIPPED') skippedTests.push({ dimension: name, checkId: check.id || `index-${index}` });
|
|
145
|
+
if (!nonEmpty(check.evidence)) errors.push(`${location}.evidence is required for applicable checks`);
|
|
146
|
+
}
|
|
147
|
+
return clone(check);
|
|
148
|
+
});
|
|
149
|
+
|
|
150
|
+
for (const requiredClass of REQUIRED_CLASSES[name] || []) {
|
|
151
|
+
if (!seenClasses.has(requiredClass)) errors.push(`${prefix}.${name} is missing declared class ${requiredClass}; mark it N/A with a reason when inapplicable`);
|
|
152
|
+
}
|
|
153
|
+
if (required === 0 || executed === 0) errors.push(`${prefix}.${name} has no executed applicable case`);
|
|
154
|
+
return {
|
|
155
|
+
applicable: true,
|
|
156
|
+
passed,
|
|
157
|
+
required,
|
|
158
|
+
percentage: required === 0 ? 0 : round(passed / required * 100),
|
|
159
|
+
checks
|
|
160
|
+
};
|
|
161
|
+
}
|
|
162
|
+
|
|
163
|
+
function evaluateDeterminism(dimension, source, errors, prefix, naDeclarations) {
|
|
164
|
+
if (!isObject(dimension) || typeof dimension.applicable !== 'boolean') {
|
|
165
|
+
errors.push(`${prefix}.determinism requires an explicit boolean applicable field`);
|
|
166
|
+
return { ...emptyDimension(true), matchingRuns: 0, totalRuns: 0, runs: [], normalizations: [] };
|
|
167
|
+
}
|
|
168
|
+
if (!dimension.applicable) {
|
|
169
|
+
if (!nonEmpty(dimension.reason)) errors.push(`${prefix}.determinism N/A requires a written reason`);
|
|
170
|
+
naDeclarations.push({ dimension: 'determinism', reason: dimension.reason || '' });
|
|
171
|
+
return { ...emptyDimension(false), matchingRuns: 0, totalRuns: 0, runs: [], normalizations: [], reason: dimension.reason || '' };
|
|
172
|
+
}
|
|
173
|
+
|
|
174
|
+
const runs = Array.isArray(dimension.runs) ? dimension.runs : [];
|
|
175
|
+
const normalizations = Array.isArray(dimension.normalizations) ? dimension.normalizations : [];
|
|
176
|
+
if (runs.length < 2) errors.push(`${prefix}.determinism requires at least two runs`);
|
|
177
|
+
if (normalizations.length > 0 && source.status !== 'AUTHORITATIVE') {
|
|
178
|
+
errors.push(`${prefix}.determinism normalization requires an authoritative source`);
|
|
179
|
+
}
|
|
180
|
+
const ids = new Set();
|
|
181
|
+
for (const [index, run] of runs.entries()) {
|
|
182
|
+
const location = `${prefix}.determinism.runs[${index}]`;
|
|
183
|
+
if (!isObject(run)) { errors.push(`${location} must be an object`); continue; }
|
|
184
|
+
if (!nonEmpty(run.id)) errors.push(`${location}.id is required`);
|
|
185
|
+
else if (ids.has(run.id)) errors.push(`${location}.id is duplicated: ${run.id}`);
|
|
186
|
+
else ids.add(run.id);
|
|
187
|
+
for (const field of ['input', 'exitStatus', 'errorCategory', 'output', 'sideEffects']) {
|
|
188
|
+
if (!Object.prototype.hasOwnProperty.call(run, field)) errors.push(`${location}.${field} is required`);
|
|
189
|
+
}
|
|
190
|
+
if (!Number.isInteger(run.exitStatus)) errors.push(`${location}.exitStatus must be an integer`);
|
|
191
|
+
if (run.errorCategory !== null && typeof run.errorCategory !== 'string') errors.push(`${location}.errorCategory must be a string or null`);
|
|
192
|
+
if (!isObject(run.sideEffects)) errors.push(`${location}.sideEffects must be an object`);
|
|
193
|
+
else {
|
|
194
|
+
for (const field of SIDE_EFFECT_FIELDS) {
|
|
195
|
+
if (!Object.prototype.hasOwnProperty.call(run.sideEffects, field)) errors.push(`${location}.sideEffects.${field} is required`);
|
|
196
|
+
}
|
|
197
|
+
for (const field of ['createdFiles', 'changedFiles', 'deletedFiles', 'events', 'logs', 'externalCalls']) {
|
|
198
|
+
if (!Array.isArray(run.sideEffects[field])) errors.push(`${location}.sideEffects.${field} must be an array`);
|
|
199
|
+
}
|
|
200
|
+
}
|
|
201
|
+
}
|
|
202
|
+
|
|
203
|
+
let matchingRuns = runs.length > 0 ? 1 : 0;
|
|
204
|
+
if (runs.length > 0) {
|
|
205
|
+
const baselineInput = runs[0].input;
|
|
206
|
+
const baseline = normalizeRun(runs[0], normalizations, errors, `${prefix}.determinism`);
|
|
207
|
+
for (let index = 1; index < runs.length; index++) {
|
|
208
|
+
const candidate = normalizeRun(runs[index], normalizations, errors, `${prefix}.determinism`);
|
|
209
|
+
if (equal(baselineInput, runs[index].input) && equal(baseline, candidate)) matchingRuns++;
|
|
210
|
+
}
|
|
211
|
+
}
|
|
212
|
+
return {
|
|
213
|
+
applicable: true,
|
|
214
|
+
passed: matchingRuns,
|
|
215
|
+
required: runs.length,
|
|
216
|
+
percentage: runs.length === 0 ? 0 : round(matchingRuns / runs.length * 100),
|
|
217
|
+
matchingRuns,
|
|
218
|
+
totalRuns: runs.length,
|
|
219
|
+
runs: clone(runs),
|
|
220
|
+
normalizations: clone(normalizations)
|
|
221
|
+
};
|
|
222
|
+
}
|
|
223
|
+
|
|
224
|
+
function evaluateSource(source, errors, prefix) {
|
|
225
|
+
if (!isObject(source)) {
|
|
226
|
+
errors.push(`${prefix}.source is required`);
|
|
227
|
+
return { source: {}, traced: 0, total: 1, percentage: 0 };
|
|
228
|
+
}
|
|
229
|
+
if (!['AUTHORITATIVE', 'INFERRED'].includes(source.status)) errors.push(`${prefix}.source.status must be AUTHORITATIVE or INFERRED`);
|
|
230
|
+
if (!nonEmpty(source.type)) errors.push(`${prefix}.source.type is required`);
|
|
231
|
+
if (source.status === 'AUTHORITATIVE') {
|
|
232
|
+
if (!nonEmpty(source.path)) errors.push(`${prefix}.source.path is required for authoritative sources`);
|
|
233
|
+
if (!nonEmpty(source.section)) errors.push(`${prefix}.source.section is required for authoritative sources`);
|
|
234
|
+
}
|
|
235
|
+
if (source.status === 'INFERRED' && !nonEmpty(source.inferenceBasis)) errors.push(`${prefix}.source.inferenceBasis is required for inferred expectations`);
|
|
236
|
+
const valid = source.status === 'AUTHORITATIVE'
|
|
237
|
+
? nonEmpty(source.type) && nonEmpty(source.path) && nonEmpty(source.section)
|
|
238
|
+
: source.status === 'INFERRED' && nonEmpty(source.type) && nonEmpty(source.inferenceBasis);
|
|
239
|
+
return { source: clone(source), traced: valid ? 1 : 0, total: 1, percentage: valid ? 100 : 0 };
|
|
240
|
+
}
|
|
241
|
+
|
|
242
|
+
function scoreDimensions(dimensions) {
|
|
243
|
+
let weighted = 0;
|
|
244
|
+
let appliedWeight = 0;
|
|
245
|
+
for (const name of DIMENSIONS) {
|
|
246
|
+
const dimension = dimensions[name];
|
|
247
|
+
if (!dimension || dimension.applicable === false || dimension.percentage === null) continue;
|
|
248
|
+
weighted += dimension.percentage * WEIGHTS[name];
|
|
249
|
+
appliedWeight += WEIGHTS[name];
|
|
250
|
+
}
|
|
251
|
+
return appliedWeight === 0 ? 0 : round(weighted / appliedWeight);
|
|
252
|
+
}
|
|
253
|
+
|
|
254
|
+
function scoringDetails(dimensions) {
|
|
255
|
+
let weightedPoints = 0;
|
|
256
|
+
let applicableWeight = 0;
|
|
257
|
+
for (const name of DIMENSIONS) {
|
|
258
|
+
const dimension = dimensions[name];
|
|
259
|
+
if (!dimension || dimension.applicable === false || dimension.percentage === null) continue;
|
|
260
|
+
weightedPoints += dimension.percentage * WEIGHTS[name];
|
|
261
|
+
applicableWeight += WEIGHTS[name];
|
|
262
|
+
}
|
|
263
|
+
return {
|
|
264
|
+
weights: { ...WEIGHTS },
|
|
265
|
+
method: 'weightedPoints / applicableWeight',
|
|
266
|
+
weightedPoints: round(weightedPoints),
|
|
267
|
+
applicableWeight: round(applicableWeight),
|
|
268
|
+
requiredScore: 100
|
|
269
|
+
};
|
|
270
|
+
}
|
|
271
|
+
|
|
272
|
+
function evaluateRequirement(requirement, index, errors) {
|
|
273
|
+
const prefix = `requirements[${index}]`;
|
|
274
|
+
const localErrorsBefore = errors.length;
|
|
275
|
+
if (!['unit', 'integration'].includes(requirement.testProfile)) {
|
|
276
|
+
errors.push(`${prefix}.testProfile must be unit or integration`);
|
|
277
|
+
}
|
|
278
|
+
if (!Array.isArray(requirement.skippedTests)) errors.push(`${prefix}.skippedTests must be an array`);
|
|
279
|
+
if (!Array.isArray(requirement.naDeclarations)) errors.push(`${prefix}.naDeclarations must be an array`);
|
|
280
|
+
if (requirement.flakyTests !== undefined && !Array.isArray(requirement.flakyTests)) errors.push(`${prefix}.flakyTests must be an array`);
|
|
281
|
+
const skippedTests = Array.isArray(requirement.skippedTests) ? clone(requirement.skippedTests) : [];
|
|
282
|
+
const naDeclarations = Array.isArray(requirement.naDeclarations) ? clone(requirement.naDeclarations) : [];
|
|
283
|
+
const sourceResult = evaluateSource(requirement.source, errors, prefix);
|
|
284
|
+
const additionalSourceResults = [];
|
|
285
|
+
if (requirement.additionalSources !== undefined && !Array.isArray(requirement.additionalSources)) {
|
|
286
|
+
errors.push(`${prefix}.additionalSources must be an array`);
|
|
287
|
+
}
|
|
288
|
+
for (const [sourceIndex, source] of (Array.isArray(requirement.additionalSources) ? requirement.additionalSources : []).entries()) {
|
|
289
|
+
additionalSourceResults.push(evaluateSource(source, errors, `${prefix}.additionalSources[${sourceIndex}]`));
|
|
290
|
+
}
|
|
291
|
+
const dimensionsInput = isObject(requirement.dimensions) ? requirement.dimensions : {};
|
|
292
|
+
if (!isObject(requirement.dimensions)) errors.push(`${prefix}.dimensions is required`);
|
|
293
|
+
const dimensions = {};
|
|
294
|
+
for (const name of DIMENSIONS.slice(0, 6)) {
|
|
295
|
+
dimensions[name] = evaluateChecks(name, dimensionsInput[name], errors, prefix, skippedTests, naDeclarations);
|
|
296
|
+
}
|
|
297
|
+
dimensions.determinism = evaluateDeterminism(dimensionsInput.determinism, requirement.source || {}, errors, prefix, naDeclarations);
|
|
298
|
+
if (requirement.testProfile === 'integration' && dimensions.determinism.applicable === false) {
|
|
299
|
+
errors.push(`${prefix}.integration profile requires applicable determinism evidence`);
|
|
300
|
+
}
|
|
301
|
+
if (!isObject(dimensionsInput.sourceTraceability) || dimensionsInput.sourceTraceability.applicable !== true) {
|
|
302
|
+
errors.push(`${prefix}.sourceTraceability must be applicable`);
|
|
303
|
+
}
|
|
304
|
+
dimensions.sourceTraceability = {
|
|
305
|
+
applicable: true,
|
|
306
|
+
passed: sourceResult.traced + additionalSourceResults.reduce((sum, result) => sum + result.traced, 0),
|
|
307
|
+
required: 1 + additionalSourceResults.length,
|
|
308
|
+
traced: sourceResult.traced + additionalSourceResults.reduce((sum, result) => sum + result.traced, 0),
|
|
309
|
+
total: 1 + additionalSourceResults.length,
|
|
310
|
+
percentage: sourceResult.traced + additionalSourceResults.reduce((sum, result) => sum + result.traced, 0) === 1 + additionalSourceResults.length ? 100 : 0
|
|
311
|
+
};
|
|
312
|
+
|
|
313
|
+
if (!nonEmpty(requirement.requirementId)) errors.push(`${prefix}.requirementId is required`);
|
|
314
|
+
if (!['PASS', 'NON_CONFORMANT'].includes(requirement.sourceConformance)) errors.push(`${prefix}.sourceConformance must be PASS or NON_CONFORMANT`);
|
|
315
|
+
if (!['VALID', 'SOURCE_DEFECT', 'SOURCE_CONFLICT'].includes(requirement.sourceValidity)) errors.push(`${prefix}.sourceValidity must be VALID, SOURCE_DEFECT, or SOURCE_CONFLICT`);
|
|
316
|
+
if (['SOURCE_DEFECT', 'SOURCE_CONFLICT'].includes(requirement.sourceValidity) && !nonEmpty(requirement.sourceNotes)) {
|
|
317
|
+
errors.push(`${prefix}.sourceNotes is required for ${requirement.sourceValidity}`);
|
|
318
|
+
}
|
|
319
|
+
if (requirement.sourceValidity === 'SOURCE_CONFLICT' && additionalSourceResults.length === 0) {
|
|
320
|
+
errors.push(`${prefix}.additionalSources must identify the conflicting authoritative source`);
|
|
321
|
+
}
|
|
322
|
+
const declaredSources = [requirement.source, ...(Array.isArray(requirement.additionalSources) ? requirement.additionalSources : [])];
|
|
323
|
+
if (requirement.sourceValidity === 'SOURCE_CONFLICT' && declaredSources.some(source => !isObject(source) || source.status !== 'AUTHORITATIVE')) {
|
|
324
|
+
errors.push(`${prefix}.SOURCE_CONFLICT may only identify authoritative sources`);
|
|
325
|
+
}
|
|
326
|
+
|
|
327
|
+
const scoring = scoringDetails(dimensions);
|
|
328
|
+
const score = scoreDimensions(dimensions);
|
|
329
|
+
const gates = {
|
|
330
|
+
applicableDimensionsExecuted: DIMENSIONS.every(name => dimensions[name].applicable === false || dimensions[name].required > 0),
|
|
331
|
+
sourceTraceabilityComplete: dimensions.sourceTraceability.percentage === 100,
|
|
332
|
+
noSourceMismatch: requirement.sourceConformance === 'PASS',
|
|
333
|
+
noSourceConflict: requirement.sourceValidity !== 'SOURCE_CONFLICT',
|
|
334
|
+
allRequiredInputOutputValidated: ['inputCompleteness', 'outputCompleteness'].every(name => dimensions[name].applicable === false || dimensions[name].percentage === 100),
|
|
335
|
+
allDeclaredErrorPathsTested: dimensions.errorHandling.applicable === false || dimensions.errorHandling.percentage === 100,
|
|
336
|
+
integrationRepeated: dimensions.determinism.applicable === false || dimensions.determinism.totalRuns >= 2,
|
|
337
|
+
deterministicOutputs: dimensions.determinism.applicable === false || dimensions.determinism.percentage === 100,
|
|
338
|
+
noSkippedOrFlakyTests: skippedTests.length === 0 && (!Array.isArray(requirement.flakyTests) || requirement.flakyTests.length === 0),
|
|
339
|
+
qualityScoreComplete: score === 100
|
|
340
|
+
};
|
|
341
|
+
const result = errors.length === localErrorsBefore && Object.values(gates).every(Boolean) ? 'PASS' : 'FAIL';
|
|
342
|
+
return {
|
|
343
|
+
requirementId: requirement.requirementId || null,
|
|
344
|
+
testProfile: requirement.testProfile || null,
|
|
345
|
+
source: sourceResult.source,
|
|
346
|
+
additionalSources: additionalSourceResults.map(result => result.source),
|
|
347
|
+
sourceConformance: requirement.sourceConformance || null,
|
|
348
|
+
sourceValidity: requirement.sourceValidity || null,
|
|
349
|
+
...(requirement.sourceNotes ? { sourceNotes: requirement.sourceNotes } : {}),
|
|
350
|
+
dimensions,
|
|
351
|
+
skippedTests,
|
|
352
|
+
flakyTests: Array.isArray(requirement.flakyTests) ? clone(requirement.flakyTests) : [],
|
|
353
|
+
naDeclarations,
|
|
354
|
+
mandatoryGates: gates,
|
|
355
|
+
scoring,
|
|
356
|
+
score,
|
|
357
|
+
result
|
|
358
|
+
};
|
|
359
|
+
}
|
|
360
|
+
|
|
361
|
+
function aggregateDimensions(requirements) {
|
|
362
|
+
const aggregate = {};
|
|
363
|
+
for (const name of DIMENSIONS) {
|
|
364
|
+
const applicable = requirements.filter(requirement => requirement.dimensions[name].applicable !== false);
|
|
365
|
+
const passed = applicable.reduce((sum, requirement) => sum + requirement.dimensions[name].passed, 0);
|
|
366
|
+
const required = applicable.reduce((sum, requirement) => sum + requirement.dimensions[name].required, 0);
|
|
367
|
+
aggregate[name] = {
|
|
368
|
+
applicable: applicable.length > 0,
|
|
369
|
+
passed,
|
|
370
|
+
required,
|
|
371
|
+
percentage: required === 0 ? null : round(passed / required * 100)
|
|
372
|
+
};
|
|
373
|
+
if (name === 'determinism') {
|
|
374
|
+
aggregate[name].matchingRuns = passed;
|
|
375
|
+
aggregate[name].totalRuns = required;
|
|
376
|
+
}
|
|
377
|
+
if (name === 'sourceTraceability') {
|
|
378
|
+
aggregate[name].traced = passed;
|
|
379
|
+
aggregate[name].total = required;
|
|
380
|
+
}
|
|
381
|
+
}
|
|
382
|
+
return aggregate;
|
|
383
|
+
}
|
|
384
|
+
|
|
385
|
+
function evaluateEvidence(evidence) {
|
|
386
|
+
const errors = [];
|
|
387
|
+
if (!isObject(evidence)) errors.push('Evidence root must be an object');
|
|
388
|
+
if (isObject(evidence) && evidence.schemaVersion !== VERSION) errors.push(`schemaVersion must be ${VERSION}`);
|
|
389
|
+
if (!isObject(evidence) || !Array.isArray(evidence.requirements) || evidence.requirements.length === 0) {
|
|
390
|
+
errors.push('requirements must be a non-empty array');
|
|
391
|
+
}
|
|
392
|
+
|
|
393
|
+
const inputRequirements = isObject(evidence) && Array.isArray(evidence.requirements) ? evidence.requirements : [];
|
|
394
|
+
const requirements = [];
|
|
395
|
+
const seenIds = new Set();
|
|
396
|
+
inputRequirements.forEach((requirement, index) => {
|
|
397
|
+
if (!isObject(requirement)) {
|
|
398
|
+
errors.push(`requirements[${index}] must be an object`);
|
|
399
|
+
return;
|
|
400
|
+
}
|
|
401
|
+
if (nonEmpty(requirement.requirementId)) {
|
|
402
|
+
if (seenIds.has(requirement.requirementId)) errors.push(`Duplicate requirementId: ${requirement.requirementId}`);
|
|
403
|
+
seenIds.add(requirement.requirementId);
|
|
404
|
+
}
|
|
405
|
+
requirements.push(evaluateRequirement(requirement, index, errors));
|
|
406
|
+
});
|
|
407
|
+
|
|
408
|
+
const dimensions = aggregateDimensions(requirements);
|
|
409
|
+
const scoring = scoringDetails(dimensions);
|
|
410
|
+
const score = scoreDimensions(dimensions);
|
|
411
|
+
const mandatoryGates = {
|
|
412
|
+
applicableDimensionsExecuted: requirements.length > 0 && requirements.every(req => req.mandatoryGates.applicableDimensionsExecuted),
|
|
413
|
+
sourceTraceabilityComplete: requirements.length > 0 && requirements.every(req => req.mandatoryGates.sourceTraceabilityComplete),
|
|
414
|
+
noSourceMismatch: requirements.length > 0 && requirements.every(req => req.mandatoryGates.noSourceMismatch),
|
|
415
|
+
noSourceConflict: requirements.length > 0 && requirements.every(req => req.mandatoryGates.noSourceConflict),
|
|
416
|
+
allRequiredInputOutputValidated: requirements.length > 0 && requirements.every(req => req.mandatoryGates.allRequiredInputOutputValidated),
|
|
417
|
+
allDeclaredErrorPathsTested: requirements.length > 0 && requirements.every(req => req.mandatoryGates.allDeclaredErrorPathsTested),
|
|
418
|
+
integrationRepeated: requirements.length > 0 && requirements.every(req => req.mandatoryGates.integrationRepeated),
|
|
419
|
+
deterministicOutputs: requirements.length > 0 && requirements.every(req => req.mandatoryGates.deterministicOutputs),
|
|
420
|
+
noSkippedOrFlakyTests: requirements.length > 0 && requirements.every(req => req.mandatoryGates.noSkippedOrFlakyTests),
|
|
421
|
+
qualityScoreComplete: score === 100
|
|
422
|
+
};
|
|
423
|
+
const skippedTests = requirements.flatMap(req => req.skippedTests.map(test => ({ requirementId: req.requirementId, ...test })));
|
|
424
|
+
const naDeclarations = requirements.flatMap(req => req.naDeclarations.map(declaration => ({ requirementId: req.requirementId, ...declaration })));
|
|
425
|
+
const result = errors.length === 0 && Object.values(mandatoryGates).every(Boolean) ? 'PASS' : 'FAIL';
|
|
426
|
+
return { schemaVersion: VERSION, requirements, dimensions, mandatoryGates, skippedTests, naDeclarations, scoring, score, result, errors };
|
|
427
|
+
}
|
|
428
|
+
|
|
429
|
+
function printSummary(report) {
|
|
430
|
+
console.log(`TDD Quality Gate: ${report.result}`);
|
|
431
|
+
console.log(`Score: ${report.score}%`);
|
|
432
|
+
console.log('Requirements:');
|
|
433
|
+
for (const requirement of report.requirements) {
|
|
434
|
+
console.log(`- ${requirement.requirementId} [${requirement.testProfile}]: ${requirement.result} ${requirement.score}%`);
|
|
435
|
+
}
|
|
436
|
+
console.log('Dimensions:');
|
|
437
|
+
for (const name of DIMENSIONS) {
|
|
438
|
+
const dimension = report.dimensions[name];
|
|
439
|
+
const score = dimension.percentage === null ? 'N/A' : `${dimension.percentage}%`;
|
|
440
|
+
console.log(`- ${name}: ${dimension.passed}/${dimension.required} (${score})`);
|
|
441
|
+
}
|
|
442
|
+
for (const [name, passed] of Object.entries(report.mandatoryGates)) console.log(`- gate ${name}: ${passed ? 'PASS' : 'FAIL'}`);
|
|
443
|
+
for (const error of report.errors) console.error(`- error: ${error}`);
|
|
444
|
+
}
|
|
445
|
+
|
|
446
|
+
function runCli(argv) {
|
|
447
|
+
const args = argv.slice(2);
|
|
448
|
+
const inputPath = args[0];
|
|
449
|
+
const outputIndex = args.indexOf('--output');
|
|
450
|
+
const outputPath = outputIndex >= 0 ? args[outputIndex + 1] : null;
|
|
451
|
+
if (!inputPath || (outputIndex >= 0 && !outputPath)) {
|
|
452
|
+
console.error('Usage: node tdd/scripts/quality-gate.js <evidence.json> [--output <report.json>]');
|
|
453
|
+
return 2;
|
|
454
|
+
}
|
|
455
|
+
let evidence;
|
|
456
|
+
try {
|
|
457
|
+
evidence = JSON.parse(fs.readFileSync(path.resolve(inputPath), 'utf8'));
|
|
458
|
+
} catch (error) {
|
|
459
|
+
console.error(`TDD Quality Gate: FAIL\n- error: ${error.message}`);
|
|
460
|
+
return 2;
|
|
461
|
+
}
|
|
462
|
+
const report = evaluateEvidence(evidence);
|
|
463
|
+
printSummary(report);
|
|
464
|
+
if (outputPath) {
|
|
465
|
+
fs.mkdirSync(path.dirname(path.resolve(outputPath)), { recursive: true });
|
|
466
|
+
fs.writeFileSync(path.resolve(outputPath), `${JSON.stringify(report, null, 2)}\n`, 'utf8');
|
|
467
|
+
} else {
|
|
468
|
+
console.log(JSON.stringify(report, null, 2));
|
|
469
|
+
}
|
|
470
|
+
return report.result === 'PASS' ? 0 : 1;
|
|
471
|
+
}
|
|
472
|
+
|
|
473
|
+
if (require.main === module) process.exit(runCli(process.argv));
|
|
474
|
+
|
|
475
|
+
module.exports = { DIMENSIONS, REQUIRED_CLASSES, VERSION, evaluateEvidence, runCli };
|
package/to-spec/SKILL.md
CHANGED
|
@@ -4,7 +4,7 @@ description: "Turn the conversation into a written spec — feature PRD, CLI/API
|
|
|
4
4
|
license: Apache-2.0
|
|
5
5
|
metadata:
|
|
6
6
|
author: Miya Daniel
|
|
7
|
-
version: 0.
|
|
7
|
+
version: 0.4.0
|
|
8
8
|
---
|
|
9
9
|
|
|
10
10
|
# To Spec
|
|
@@ -22,15 +22,13 @@ Turns the conversation into one published spec artifact. No interviewing — `gr
|
|
|
22
22
|
|
|
23
23
|
## Workflow
|
|
24
24
|
|
|
25
|
-
1. Resolve path via `node "
|
|
25
|
+
1. Resolve path via the shared `<skills-repo-root>/multi-agent-workspace/scripts/project-docs-resolver.js` contract and `node "<this-skill-dir>/scripts/check-project-docs.js" check`; use explicit projectDocs, CONTEXT-MAP, inferred docs, then committable fallback.
|
|
26
26
|
2. Zero-trust context: cite `Evidence: <file:line> -> Finding: <meaning>`; blocking forks mean run `grill-me`/`grill-with-docs` first.
|
|
27
27
|
3. Mandatory outline preview (closest fit, 10-20 lines + target path); write only after user confirms.
|
|
28
28
|
4. Publish to `specs/<feature-slug>.md` (`Status: ready-for-agent`) or `reference/`, `adr/`.
|
|
29
29
|
5. Golden Flow: Feature spec needs a Design Audit (`multi-agent-workspace`) before `/to-tickets`.
|
|
30
30
|
|
|
31
|
-
|
|
32
|
-
|
|
33
|
-
Deep dive: references/process.md
|
|
31
|
+
Deep dive: <this-skill-dir>/references/process.md
|
|
34
32
|
|
|
35
33
|
## USE FOR:
|
|
36
34
|
- conversation into a published PRD or design doc
|
|
@@ -37,15 +37,10 @@
|
|
|
37
37
|
*/
|
|
38
38
|
const fs = require('fs');
|
|
39
39
|
const path = require('path');
|
|
40
|
-
|
|
41
|
-
|
|
42
|
-
|
|
43
|
-
|
|
44
|
-
if (fs.existsSync(path.join(dir, '.git'))) return dir;
|
|
45
|
-
dir = path.dirname(dir);
|
|
46
|
-
}
|
|
47
|
-
return process.cwd();
|
|
48
|
-
}
|
|
40
|
+
// Keep project-doc path decisions in the resolver carried by the installed
|
|
41
|
+
// multi-agent-workspace skill. This script consumes the same contract.
|
|
42
|
+
const { getWorkspaceRoot, getRepoManifestHomes, detectProjectDocsConventions } =
|
|
43
|
+
require('./project-docs-resolver');
|
|
49
44
|
|
|
50
45
|
function loadManifestHelper() {
|
|
51
46
|
const candidates = [
|
|
@@ -62,7 +57,26 @@ function loadManifestHelper() {
|
|
|
62
57
|
}
|
|
63
58
|
} catch (e) {}
|
|
64
59
|
}
|
|
65
|
-
|
|
60
|
+
// A standalone installed skill may not carry the package helper. Keep its
|
|
61
|
+
// manifest format compatible with the package's tiny helper instead of
|
|
62
|
+
// making document resolution depend on the source checkout.
|
|
63
|
+
return {
|
|
64
|
+
getManifestPath: home => path.join(home, 'harness-everything', 'manifest.json'),
|
|
65
|
+
readManifest: file => {
|
|
66
|
+
try {
|
|
67
|
+
const data = JSON.parse(fs.readFileSync(file, 'utf8'));
|
|
68
|
+
if (!Array.isArray(data.skills)) data.skills = [];
|
|
69
|
+
if (!Array.isArray(data.agents)) data.agents = [];
|
|
70
|
+
return data;
|
|
71
|
+
} catch {
|
|
72
|
+
return { package: 'harness-everything', skills: [], agents: [] };
|
|
73
|
+
}
|
|
74
|
+
},
|
|
75
|
+
writeManifest: (file, data) => {
|
|
76
|
+
fs.mkdirSync(path.dirname(file), { recursive: true });
|
|
77
|
+
fs.writeFileSync(file, JSON.stringify(data, null, 2), 'utf8');
|
|
78
|
+
}
|
|
79
|
+
};
|
|
66
80
|
}
|
|
67
81
|
const { getManifestPath, readManifest, writeManifest } = loadManifestHelper();
|
|
68
82
|
|
|
@@ -78,62 +92,13 @@ const FLAG_TO_FIELD = {
|
|
|
78
92
|
// issue template or a multi-paragraph doc-location policy inline.
|
|
79
93
|
const MAX_FIELD_LENGTH = 200;
|
|
80
94
|
|
|
81
|
-
// Workspace-relative platform homes ONLY - no userHome/.agents or
|
|
82
|
-
// userHome/.claude here (see file header). tier-router.js and
|
|
83
|
-
// register-dynamic-skill.js include those two because self-evolved skills
|
|
84
|
-
// are meant to travel with the user across projects; projectDocs must not.
|
|
85
|
-
function getRepoManifestHomes(workspaceRoot) {
|
|
86
|
-
return [
|
|
87
|
-
path.join(workspaceRoot, '.claude'),
|
|
88
|
-
path.join(workspaceRoot, '.cursor'),
|
|
89
|
-
path.join(workspaceRoot, '.github'),
|
|
90
|
-
path.join(workspaceRoot, '.codex'),
|
|
91
|
-
path.join(workspaceRoot, '.continue'),
|
|
92
|
-
];
|
|
93
|
-
}
|
|
94
|
-
|
|
95
95
|
function missingFields(projectDocs) {
|
|
96
96
|
if (!projectDocs) return REQUIRED_FIELDS.slice();
|
|
97
97
|
return REQUIRED_FIELDS.filter(f => !String(projectDocs[f] || '').trim());
|
|
98
98
|
}
|
|
99
99
|
|
|
100
|
-
/**
|
|
101
|
-
* Heuristic workspace convention detector.
|
|
102
|
-
* Automatically infers docLocation, tracker, and issueDefinition from standard directory layouts.
|
|
103
|
-
*/
|
|
104
100
|
function detectWorkspaceConventions(workspaceRoot) {
|
|
105
|
-
|
|
106
|
-
|
|
107
|
-
// Detect docLocation
|
|
108
|
-
const docCandidates = ['docs/roadmaps', 'docs/reference', 'docs/specs', 'docs', 'doc'];
|
|
109
|
-
for (const dir of docCandidates) {
|
|
110
|
-
if (fs.existsSync(path.join(workspaceRoot, dir))) {
|
|
111
|
-
inferred.docLocation = `${dir}/`;
|
|
112
|
-
break;
|
|
113
|
-
}
|
|
114
|
-
}
|
|
115
|
-
|
|
116
|
-
// Detect tracker
|
|
117
|
-
if (fs.existsSync(path.join(workspaceRoot, '.github/ISSUE_TEMPLATE'))) {
|
|
118
|
-
inferred.tracker = 'GitHub Issues via .github/ISSUE_TEMPLATE/';
|
|
119
|
-
} else if (fs.existsSync(path.join(workspaceRoot, 'docs/roadmaps'))) {
|
|
120
|
-
inferred.tracker = 'docs/roadmaps/ local markdown tickets';
|
|
121
|
-
} else if (fs.existsSync(path.join(workspaceRoot, 'tasks/tickets'))) {
|
|
122
|
-
inferred.tracker = 'tasks/tickets/ local markdown tickets';
|
|
123
|
-
} else if (fs.existsSync(path.join(workspaceRoot, '.scratch'))) {
|
|
124
|
-
inferred.tracker = '.scratch/ local markdown workspace';
|
|
125
|
-
} else if (inferred.docLocation) {
|
|
126
|
-
inferred.tracker = `${inferred.docLocation}tickets/ local markdown tickets`;
|
|
127
|
-
}
|
|
128
|
-
|
|
129
|
-
// Detect issueDefinition
|
|
130
|
-
if (inferred.tracker && inferred.tracker.includes('ISSUE_TEMPLATE')) {
|
|
131
|
-
inferred.issueDefinition = '.github/ISSUE_TEMPLATE/ template structure';
|
|
132
|
-
} else if (inferred.tracker) {
|
|
133
|
-
inferred.issueDefinition = 'Title + acceptance criteria + Status: ready-for-agent';
|
|
134
|
-
}
|
|
135
|
-
|
|
136
|
-
return inferred;
|
|
101
|
+
return detectProjectDocsConventions(workspaceRoot);
|
|
137
102
|
}
|
|
138
103
|
|
|
139
104
|
function runCheck(workspaceRoot) {
|
|
@@ -247,12 +212,27 @@ Repo-local only: writes only to workspace-relative manifest homes (.claude/, .cu
|
|
|
247
212
|
}
|
|
248
213
|
|
|
249
214
|
const command = process.argv[2] || 'check';
|
|
250
|
-
const
|
|
215
|
+
const commandArgs = process.argv.slice(3);
|
|
216
|
+
const workspaceIndex = commandArgs.indexOf('--workspace');
|
|
217
|
+
// An explicit workspace is authoritative. Only the omitted flag walks up to
|
|
218
|
+
// the repository root; otherwise a nested target would silently resolve to
|
|
219
|
+
// its containing checkout and inspect the wrong project configuration.
|
|
220
|
+
if (workspaceIndex !== -1 && !commandArgs[workspaceIndex + 1]) {
|
|
221
|
+
console.error('[Project Docs Check] --workspace requires a path');
|
|
222
|
+
process.exit(1);
|
|
223
|
+
}
|
|
224
|
+
const workspaceRoot = workspaceIndex === -1
|
|
225
|
+
? getWorkspaceRoot()
|
|
226
|
+
: path.resolve(commandArgs[workspaceIndex + 1]);
|
|
227
|
+
if (!workspaceRoot) {
|
|
228
|
+
console.error('[to-spec/check-project-docs] Cannot inspect project docs without a resolved git workspace.');
|
|
229
|
+
process.exit(1);
|
|
230
|
+
}
|
|
251
231
|
|
|
252
232
|
if (command === 'check') {
|
|
253
233
|
runCheck(workspaceRoot);
|
|
254
234
|
} else if (command === 'init') {
|
|
255
|
-
runInit(workspaceRoot,
|
|
235
|
+
runInit(workspaceRoot, commandArgs);
|
|
256
236
|
} else {
|
|
257
237
|
console.error(`[Project Docs Check] Unknown command "${command}". Use 'check' or 'init'.`);
|
|
258
238
|
process.exit(1);
|