@jterrats/open-orchestra 1.0.2 → 1.0.4

package/docs/reports/ac-failure-evidence-20260517.md

@@ -0,0 +1,115 @@
+# Acceptance Criteria Failure Evidence
+
+Date: 2026-05-17
+Task: AC-FAILURE-EVIDENCE-20260517
+
+Scope: failure evidence only. This report does not remediate the failures.
+
+## Current Test Failure
+
+Command:
+
+```sh
+node --test test/docs-adoption.test.js
+```
+
+Result:
+
+- 2 passing
+- 1 failing
+
+Failing test:
+
+- `test/docs-adoption.test.js:51`
+- Test name: `first-run onboarding defines visible value before production governance`
+
+Expected:
+
+- `README.md`
+- `docs/adoption-guide.md`
+- `docs/core-command-surface.md`
+- `site/src/App.jsx`
+
+must each match:
+
+```text
+/First visible value/i
+```
+
+Observed:
+
+- `README.md` contains `First Visible Value`.
+- `docs/adoption-guide.md` contains `First visible value`.
+- `docs/core-command-surface.md` contains `first visible value`.
+- `site/src/App.jsx` does not contain that literal text anymore.
+- The site content now comes from generated data:
+  - `site/src/generated-site-content.js` contains `First visible value in minutes`.
+  - `docs/site-manifest.json` contains `First Visible Value`.
+
+Probable cause:
+
+- The site was changed from hardcoded page copy to generated content, but the test still validates literal copy inside `site/src/App.jsx`.
+
+Impact:
+
+- `npm test` fails.
+- `orchestra release check --json` fails its `tests` check.
+- The release readiness subcheck still passes, but the full release check does not.
+
+## Historical AC Traceability Failure
+
+Source report:
+
+- `docs/reports/ac-history-dry-run-20260517.md`
+
+Dry-run result:
+
+- Historical completed tasks evaluated: 77
+- Acceptance criteria evaluated: 324
+- Cumple: 8
+- No cumple: 316
+
+Coverage status breakdown:
+
+- `covered`: 8
+- `planned`: 267
+- `gap`: 20
+- `skipped`: 29
+
+Interpretation:
+
+- Most historical acceptance criteria are not currently machine-verifiable through linked evidence.
+- This does not prove every behavior is absent.
+- It does prove the repository cannot currently demonstrate AC compliance for most historical work using the current evidence model.
+
+## Explicit Gap Examples
+
+The dry-run identified these explicit `gap` criteria among others:
+
+- `WEB-E2E`: npm script runs locally.
+- `SETUP-TASK-CONTRACT`: contract includes backlog story, AC, DoR, DoD, dependencies, risks, and acceptance status.
+- `PORTABILITY`: validate workflow runs without Unix shell assumptions and CI covers cross-platform smoke paths.
+- `QA-RELEASE`: release check reports uncovered AC, blocker reviews, active locks, and missing rollback/smoke evidence.
+- `UPGRADE-DOGFOOD`: upgrade workflow can run a post-install smoke check.
+- `OO-210`: redo/sign-off workflow mode observation contract.
+- `OO-211`: deployment risk challenge and release readiness surfacing.
+- `OO-225`: default initialization creates focused playbooks for active roles/phases.
+- `GH-237`: runtime adapter setup examples and documented permission/configuration model.
+- `GH-298`: dashboard partial API failure handling and E2E proof.
+- `GH-295`: release check discrepancy and historical uncovered AC scoping.
+
+## Reproducible Commands
+
+```sh
+node --test test/docs-adoption.test.js
+npm test
+node ./bin/orchestra.js release check --json
+node ./bin/orchestra.js qa coverage --task WEB-E2E --json
+```
+
+## Recommended Remediation Order
+
+1. Fix the current `docs-adoption` test drift caused by generated site content.
+2. Re-run `npm test` and `orchestra release check --json`.
+3. Convert the 20 explicit `gap` criteria into remediation tasks or accepted-risk records.
+4. Backfill linked evidence for historical criteria that are already implemented but only show as `planned` or `skipped`.

package/docs/reports/ac-history-dry-run-20260517.md

@@ -0,0 +1,434 @@
+# Historical Acceptance Criteria Dry Run
+
+Date: 2026-05-17
+Task: AC-HISTORY-DRY-RUN-20260517
+
+Scope: completed top-level tasks with acceptance criteria. Pending, blocked, canceled, generated phase subtasks, and future backlog were excluded.
+
+Interpretation: Cumple means the current QA coverage dry-run found explicit linked evidence for the criterion. No cumple means the criterion is only planned, is a gap, or has no machine-verifiable evidence linked in the current repository state.
+
+## Summary
+
+- Historical tasks evaluated: 77
+- Acceptance criteria evaluated: 324
+- Cumple: 8
+- No cumple: 316
+- Coverage statuses: planned=267, gap=20, skipped=29, covered=8
+
+## Current Validation Commands
+
+- `npm run build`: passed
+- `npm run lint`: passed
+- `node ./bin/orchestra.js release check --json`: failed because `npm test` failed and working tree is dirty; release readiness subcheck passed.
+- `npm test`: failed with 578 passing, 1 failing, 4 skipped. Failing test: `test/docs-adoption.test.js` expects `First visible value` in `site/src/App.jsx`.
+
+## Task-Level Result
+
+| Task | Result | Cumple | No cumple |
+| --- | --- | ---: | ---: |
+| WEB-E2E | No cumple | 0 | 7 |
+| SETUP-BRIDGE | No cumple | 0 | 5 |
+| WEB-LIFECYCLE | No cumple | 0 | 11 |
+| ADVISORY-ARTIFACTS | No cumple | 0 | 5 |
+| MCP-OAUTH-PARITY | No cumple | 0 | 6 |
+| TRACEABILITY-DOCS | No cumple | 0 | 7 |
+| COMMAND-CONTRACTS | No cumple | 0 | 4 |
+| SETUP-TASK-CONTRACT | No cumple | 0 | 4 |
+| VSCODE-DEEPLINK | No cumple | 0 | 5 |
+| CONTEXT-BUDGET | No cumple | 0 | 5 |
+| INIT-HELP | No cumple | 0 | 1 |
+| INIT-CONTEXT | No cumple | 0 | 1 |
+| INIT-NEXT | No cumple | 0 | 1 |
+| INIT-SAFETY | No cumple | 0 | 1 |
+| INIT-DOGFOOD | No cumple | 0 | 1 |
+| PORTABILITY | No cumple | 0 | 1 |
+| ONBOARDING-E2E | No cumple | 0 | 1 |
+| WEB-OPS | No cumple | 0 | 1 |
+| GITHUB-SYNC | No cumple | 0 | 6 |
+| MEMORY-GOVERNANCE | No cumple | 0 | 6 |
+| QA-RELEASE | No cumple | 0 | 6 |
+| UPGRADE-DOGFOOD | No cumple | 0 | 6 |
+| DOCS-PERSONAS | No cumple | 0 | 6 |
+| ADVISORY-MODE | No cumple | 0 | 11 |
+| EVENT-VALIDATOR | No cumple | 0 | 1 |
+| ARCH-DIAGRAMS | No cumple | 0 | 1 |
+| OO-208 | No cumple | 0 | 6 |
+| OO-212 | No cumple | 0 | 4 |
+| OO-211 | No cumple | 0 | 5 |
+| OO-210 | No cumple | 0 | 6 |
+| OO-209 | No cumple | 0 | 6 |
+| OO-224 | No cumple | 0 | 13 |
+| OO-225 | No cumple | 0 | 6 |
+| OO-226 | No cumple | 0 | 6 |
+| OO-227 | No cumple | 0 | 6 |
+| OO-228 | No cumple | 0 | 6 |
+| OO-238 | No cumple | 0 | 1 |
+| GH-228 | No cumple | 0 | 5 |
+| GH-229 | No cumple | 1 | 6 |
+| GH-230 | No cumple | 2 | 3 |
+| GH-231 | No cumple | 0 | 7 |
+| GH-232 | No cumple | 0 | 5 |
+| GH-233 | No cumple | 0 | 5 |
+| GH-234 | No cumple | 0 | 5 |
+| GH-235 | No cumple | 0 | 2 |
+| GH-236 | No cumple | 1 | 4 |
+| GH-237 | No cumple | 0 | 2 |
+| GH-292 | No cumple | 0 | 1 |
+| GH-293 | No cumple | 0 | 1 |
+| RUN-VS-WORKFLOW | No cumple | 0 | 1 |
+| GH-294 | No cumple | 0 | 6 |
+| GH-298 | No cumple | 0 | 6 |
+| GH-297 | No cumple | 0 | 6 |
+| GH-295 | No cumple | 0 | 6 |
+| GH-296 | No cumple | 0 | 6 |
+| GH-299 | No cumple | 0 | 1 |
+| GH-300 | No cumple | 0 | 3 |
+| UX-UI-005 | No cumple | 0 | 1 |
+| UX-UI-007 | No cumple | 0 | 4 |
+| UX-UI-008 | No cumple | 0 | 1 |
+| UX-UI-009 | No cumple | 0 | 1 |
+| UX-UI-010 | No cumple | 0 | 1 |
+| UX-UI-011 | No cumple | 0 | 1 |
+| UX-UI-012 | No cumple | 0 | 1 |
+| ESTIMATE-BENCHMARK-CALIBRATION | No cumple | 0 | 6 |
+| DEV-STANDARDS-GENERALIZE | No cumple | 0 | 5 |
+| VISUAL-VALIDATION-GATE | No cumple | 0 | 8 |
+| SETUP-IMPORT-ROLE-ALIASES | No cumple | 0 | 4 |
+| RULES-PLAYBOOKS-HEALTH | No cumple | 0 | 5 |
+| WORKFLOW-DOCS-PLAYBOOKS-GATES | No cumple | 0 | 5 |
+| PLAYBOOK-SCAFFOLD | No cumple | 0 | 6 |
+| QA-PLAN-READY-GATE | No cumple | 0 | 1 |
+| QA-AUTOMATION-ASSERTIONS | No cumple | 0 | 1 |
+| GH-331 | No cumple | 0 | 1 |
+| GH-332 | Cumple | 4 | 0 |
+| E2E-AC-RESULTS-HARDENING | No cumple | 0 | 4 |
+| ARCH-SALESFORCE-INTEGRATION-DIAGRAM | No cumple | 0 | 4 |
+
+## Criterion-Level Result
+
+| Task | Result | Coverage | Acceptance criterion | Reason |
+| --- | --- | --- | --- | --- |
+| WEB-E2E | No cumple | planned | real web server starts | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-E2E | No cumple | planned | no mocks or route injection | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-E2E | No cumple | planned | task create persists | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-E2E | No cumple | planned | cost dashboard renders real provenance | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-E2E | No cumple | planned | Playwright evidence persists | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-E2E | No cumple | gap | npm script runs locally | gap de automatizacion/evidencia |
+| WEB-E2E | No cumple | planned | test structure reusable by QA SDET generated Playwright specs from future handoffs | solo planificado; sin evidencia ejecutada enlazada |
+| SETUP-BRIDGE | No cumple | planned | imports profiles idempotently | solo planificado; sin evidencia ejecutada enlazada |
+| SETUP-BRIDGE | No cumple | planned | maps tasks evidence handoffs | solo planificado; sin evidencia ejecutada enlazada |
+| SETUP-BRIDGE | No cumple | planned | reports conflicts | solo planificado; sin evidencia ejecutada enlazada |
+| SETUP-BRIDGE | No cumple | planned | preserves ownership boundary | solo planificado; sin evidencia ejecutada enlazada |
+| SETUP-BRIDGE | No cumple | planned | documents flow | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | create or select refined task | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | start workflow run | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | show phase and gate state | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | resume paused run | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | approve or reject gates | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | developer-to-QA handoff includes changed files commands evidence gaps and recommended Playwright coverage | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | QA SDET generates or updates Playwright plan from acceptance criteria | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | QA SDET creates or updates executable Playwright specs when browser-real validation is required | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | release gate blocks UI AC without Playwright evidence or accepted risk | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | persist workflow state | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-LIFECYCLE | No cumple | planned | E2E validates real lifecycle | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-ARTIFACTS | No cumple | planned | init advisory avoids root files | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-ARTIFACTS | No cumple | planned | generates advisory README and role guides | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-ARTIFACTS | No cumple | planned | records portable decisions | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-ARTIFACTS | No cumple | planned | exports project task | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-ARTIFACTS | No cumple | planned | unsafe roots blocked | solo planificado; sin evidencia ejecutada enlazada |
+| MCP-OAUTH-PARITY | No cumple | planned | RFC8414 discovery | solo planificado; sin evidencia ejecutada enlazada |
+| MCP-OAUTH-PARITY | No cumple | planned | RFC7591 dynamic registration | solo planificado; sin evidencia ejecutada enlazada |
+| MCP-OAUTH-PARITY | No cumple | planned | stdio proxy injects bearer | solo planificado; sin evidencia ejecutada enlazada |
+| MCP-OAUTH-PARITY | No cumple | planned | refresh before expiry | solo planificado; sin evidencia ejecutada enlazada |
+| MCP-OAUTH-PARITY | No cumple | planned | retry once on 401 | solo planificado; sin evidencia ejecutada enlazada |
+| MCP-OAUTH-PARITY | No cumple | planned | no token logging | solo planificado; sin evidencia ejecutada enlazada |
+| TRACEABILITY-DOCS | No cumple | planned | docs cover PO refinement | solo planificado; sin evidencia ejecutada enlazada |
+| TRACEABILITY-DOCS | No cumple | planned | developer delivery | solo planificado; sin evidencia ejecutada enlazada |
+| TRACEABILITY-DOCS | No cumple | planned | QA verification | solo planificado; sin evidencia ejecutada enlazada |
+| TRACEABILITY-DOCS | No cumple | planned | advisory conversion | solo planificado; sin evidencia ejecutada enlazada |
+| TRACEABILITY-DOCS | No cumple | planned | release readiness | solo planificado; sin evidencia ejecutada enlazada |
+| TRACEABILITY-DOCS | No cumple | planned | artifact traceability | solo planificado; sin evidencia ejecutada enlazada |
+| TRACEABILITY-DOCS | No cumple | planned | generated reference separation | solo planificado; sin evidencia ejecutada enlazada |
+| COMMAND-CONTRACTS | No cumple | planned | JSON commands have schemas | solo planificado; sin evidencia ejecutada enlazada |
+| COMMAND-CONTRACTS | No cumple | planned | manifest links schemas/messages | solo planificado; sin evidencia ejecutada enlazada |
+| COMMAND-CONTRACTS | No cumple | planned | fixtures validate output | solo planificado; sin evidencia ejecutada enlazada |
+| COMMAND-CONTRACTS | No cumple | planned | compatibility documented | solo planificado; sin evidencia ejecutada enlazada |
+| SETUP-TASK-CONTRACT | No cumple | gap | contract includes backlog story AC DoR DoD dependencies risks acceptance status | gap de automatizacion/evidencia |
+| SETUP-TASK-CONTRACT | No cumple | planned | old records readable | solo planificado; sin evidencia ejecutada enlazada |
+| SETUP-TASK-CONTRACT | No cumple | planned | evidence IDs preserved | solo planificado; sin evidencia ejecutada enlazada |
+| SETUP-TASK-CONTRACT | No cumple | planned | version included | solo planificado; sin evidencia ejecutada enlazada |
+| VSCODE-DEEPLINK | No cumple | planned | detects setup-agents artifacts | solo planificado; sin evidencia ejecutada enlazada |
+| VSCODE-DEEPLINK | No cumple | planned | guides from setup to init/status/workflow | solo planificado; sin evidencia ejecutada enlazada |
+| VSCODE-DEEPLINK | No cumple | planned | shows import pending state | solo planificado; sin evidencia ejecutada enlazada |
+| VSCODE-DEEPLINK | No cumple | planned | works without setup-agents | solo planificado; sin evidencia ejecutada enlazada |
+| VSCODE-DEEPLINK | No cumple | planned | docs boundary | solo planificado; sin evidencia ejecutada enlazada |
+| CONTEXT-BUDGET | No cumple | planned | context reports estimatedTokens and tokenBudget | solo planificado; sin evidencia ejecutada enlazada |
+| CONTEXT-BUDGET | No cumple | planned | context supports --budget for JSON and human output | solo planificado; sin evidencia ejecutada enlazada |
+| CONTEXT-BUDGET | No cumple | planned | runtime brief and workflow render honor bounded context | solo planificado; sin evidencia ejecutada enlazada |
+| CONTEXT-BUDGET | No cumple | planned | large evidence handoff review history is trimmed by priority | solo planificado; sin evidencia ejecutada enlazada |
+| CONTEXT-BUDGET | No cumple | planned | tests prove high-history tasks stay under budget | solo planificado; sin evidencia ejecutada enlazada |
+| INIT-HELP | No cumple | planned | init --help succeeds in initialized workspace; help includes project advisory target-dir target runtime force and confirm-unknown examples; help writes no files; root help points to recommended first-use path; dogfood evidence uses installed orchestra binary | solo planificado; sin evidencia ejecutada enlazada |
+| INIT-CONTEXT | No cumple | planned | follow-up commands can validate the same target dir initialized by init; health status validate and task list do not silently report caller repo when a target root is requested; human and JSON outputs expose effective root where needed; cwd default behavior is unchanged; dogfood covers temp initialized workspace | solo planificado; sin evidencia ejecutada enlazada |
+| INIT-NEXT | No cumple | planned | project init prints health task and workflow/web next steps; advisory init prints advisory-specific next steps; existing workspace failure explains status health force and target-dir options; unknown workspace guidance explains confirm-unknown advisory and target-dir; structured output exists or is documented as follow-up | solo planificado; sin evidencia ejecutada enlazada |
+| INIT-SAFETY | No cumple | planned | init reports existing managed artifacts and preserve/overwrite behavior; force mode lists generated files refreshed; dry-run or check path writes nothing; conflicts include remediation; safety docs cover project and advisory modes | solo planificado; sin evidencia ejecutada enlazada |
+| INIT-DOGFOOD | No cumple | planned | e2e uses orchestra from PATH not node bin/orchestra.js; covers project init advisory init unknown confirmation existing workspace and follow-up health/status; uses real temp dirs with no mocks; CI or release gate can run suite; release readiness records evidence | solo planificado; sin evidencia ejecutada enlazada |
+| PORTABILITY | No cumple | gap | validate:workflow runs without Unix sh; CI covers cross-platform smoke paths; docs state supported platforms and limitations; macOS and Linux behavior remains unchanged | gap de automatizacion/evidencia |
+| ONBOARDING-E2E | No cumple | planned | real temp workspace; no mocks; uses installed orchestra binary; documents happy path; failure output includes recovery steps | solo planificado; sin evidencia ejecutada enlazada |
+| WEB-OPS | No cumple | planned | shows tasks owners statuses blockers gates evidence and next actions; task detail includes AC DoR DoD risks handoffs reviews evidence release readiness; mobile desktop responsive; Playwright no-mock coverage | solo planificado; sin evidencia ejecutada enlazada |
+| GITHUB-SYNC | No cumple | planned | sync imports issue metadata; status/evidence comments are idempotent; closure requires DoD evidence or accepted risk; mapping documented and configurable | solo planificado; sin evidencia ejecutada enlazada |
+| GITHUB-SYNC | No cumple | planned | Command can import/sync GitHub issue metadata into Orchestra tasks. | solo planificado; sin evidencia ejecutada enlazada |
+| GITHUB-SYNC | No cumple | planned | Command can update GitHub issue comments with evidence and status. | solo planificado; sin evidencia ejecutada enlazada |
+| GITHUB-SYNC | No cumple | planned | Closure requires passing DoD evidence or explicit accepted risk. | solo planificado; sin evidencia ejecutada enlazada |
+| GITHUB-SYNC | No cumple | planned | Labels/milestones/status mapping is documented and configurable. | solo planificado; sin evidencia ejecutada enlazada |
+| GITHUB-SYNC | No cumple | planned | Failures are idempotent and do not duplicate comments/tasks. | solo planificado; sin evidencia ejecutada enlazada |
+| MEMORY-GOVERNANCE | No cumple | planned | role bundles for PO architect developer QA release; included/omitted sections visible; token budget and estimates shown; high-history tests stay bounded; no raw secrets or provider responses | solo planificado; sin evidencia ejecutada enlazada |
+| MEMORY-GOVERNANCE | No cumple | planned | Memory output explains included and omitted sections. | solo planificado; sin evidencia ejecutada enlazada |
+| MEMORY-GOVERNANCE | No cumple | planned | Bundles are role-scoped for PO, architect, developer, QA, and release. | solo planificado; sin evidencia ejecutada enlazada |
+| MEMORY-GOVERNANCE | No cumple | planned | Lessons learned and prompt registry entries are retrieved through explicit hooks. | solo planificado; sin evidencia ejecutada enlazada |
+| MEMORY-GOVERNANCE | No cumple | planned | Token budget, estimated tokens, and trimming rationale are visible in JSON and human output. | solo planificado; sin evidencia ejecutada enlazada |
+| MEMORY-GOVERNANCE | No cumple | planned | Tests prove high-history workspaces stay within budget without dropping critical AC. | solo planificado; sin evidencia ejecutada enlazada |
+| QA-RELEASE | No cumple | planned | AC maps to unit integration e2e manual or accepted-risk evidence; release check reports uncovered AC blockers locks smoke rollback gaps; browser AC requires Playwright evidence or accepted risk | solo planificado; sin evidencia ejecutada enlazada |
+| QA-RELEASE | No cumple | planned | Each acceptance criterion can be mapped to unit, integration, E2E, manual, or accepted-risk evidence. | solo planificado; sin evidencia ejecutada enlazada |
+| QA-RELEASE | No cumple | gap | Release check reports uncovered AC, unresolved blocker reviews, active locks, and missing rollback/smoke evidence. | gap de automatizacion/evidencia |
+| QA-RELEASE | No cumple | planned | Browser/UI AC requires Playwright evidence or accepted risk. | solo planificado; sin evidencia ejecutada enlazada |
+| QA-RELEASE | No cumple | planned | Output is available as JSON and human-readable summary. | solo planificado; sin evidencia ejecutada enlazada |
+| QA-RELEASE | No cumple | planned | Web console can display release readiness status. | solo planificado; sin evidencia ejecutada enlazada |
+| UPGRADE-DOGFOOD | No cumple | planned | detects latest beta versions; prints exact canonical install command; post-install smoke check available; health shows installed root/version; docs explain stable vs beta | solo planificado; sin evidencia ejecutada enlazada |
+| UPGRADE-DOGFOOD | No cumple | planned | Command or health output detects latest/beta versions and prints exact install command. | solo planificado; sin evidencia ejecutada enlazada |
+| UPGRADE-DOGFOOD | No cumple | gap | Upgrade workflow can run a post-install smoke check. | gap de automatizacion/evidencia |
+| UPGRADE-DOGFOOD | No cumple | planned | Installed package root/version are visible in health output. | solo planificado; sin evidencia ejecutada enlazada |
+| UPGRADE-DOGFOOD | No cumple | planned | Dogfood suite can verify package contents and CLI binary from PATH. | solo planificado; sin evidencia ejecutada enlazada |
+| UPGRADE-DOGFOOD | No cumple | planned | Docs explain stable vs beta installation. | solo planificado; sin evidencia ejecutada enlazada |
+| DOCS-PERSONAS | No cumple | planned | persona docs include commands artifacts blockers recovery; docs link to current product flows; site reflects implemented capabilities; canonical package and CLI names used | solo planificado; sin evidencia ejecutada enlazada |
+| DOCS-PERSONAS | No cumple | planned | Docs include paths for PO refinement, developer execution, QA validation, tech lead oversight, and release management. | solo planificado; sin evidencia ejecutada enlazada |
+| DOCS-PERSONAS | No cumple | planned | Each path includes commands, expected artifacts, common blockers, and recovery steps. | solo planificado; sin evidencia ejecutada enlazada |
+| DOCS-PERSONAS | No cumple | planned | Docs link to first-use onboarding, web console, memory, GitHub sync, and release readiness. | solo planificado; sin evidencia ejecutada enlazada |
+| DOCS-PERSONAS | No cumple | planned | Public site reflects the same current product capabilities. | solo planificado; sin evidencia ejecutada enlazada |
+| DOCS-PERSONAS | No cumple | planned | Docs avoid stale package names and use installed CLI form. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | advisory init creates clear artifacts decisions role guides conversion path; findings can promote to project tasks; mode visible in health/status; unsafe roots blocked; docs explain advisory vs project mode | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Advisory init creates clear artifacts, decisions, role guides, and conversion path. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Advisory findings can be promoted into real project tasks. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Advisory mode is visible in health/status output. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Unsafe roots remain blocked, and advisory writes stay inside workflow-owned paths unless explicitly targeted. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Docs explain when to use advisory vs project mode. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Advisory init creates clear artifacts, decisions, role guides, and conversion path. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Advisory findings can be promoted into real project tasks. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Advisory mode is visible in health/status output. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Unsafe roots remain blocked, and advisory writes stay inside workflow-owned paths unless explicitly targeted. | solo planificado; sin evidencia ejecutada enlazada |
+| ADVISORY-MODE | No cumple | planned | Docs explain when to use advisory vs project mode. | solo planificado; sin evidencia ejecutada enlazada |
+| EVENT-VALIDATOR | No cumple | planned | ESTIMATE_RECORDED is accepted; estimate then validate has no unknown event type warning; tests cover estimate/validate path; emitted event catalog drift is reviewed | solo planificado; sin evidencia ejecutada enlazada |
+| ARCH-DIAGRAMS | No cumple | planned | Site and docs include architecture diagrams; diagrams are readable in light/dark contexts; build and precommit pass | solo planificado; sin evidencia ejecutada enlazada |
+| OO-208 | No cumple | planned | Workflow configuration can insert a challenge role between Architect and Developer without hardcoding a single domain-specific role. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-208 | No cumple | planned | The default challenge role can be reviewer_critic, domain_expert, platform_engineer, or a configured project role. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-208 | No cumple | planned | Phase-mode gates can pause at the new architecture challenge transition for human triage. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-208 | No cumple | planned | Challenge output is stored as a workflow artifact and included in downstream phase context. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-208 | No cumple | planned | If the challenge phase blocks the design, the workflow routes back to Architect or stops with a clear recovery command. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-208 | No cumple | planned | Existing six-phase workflows continue to work when no challenge phase is configured. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-212 | No cumple | skipped | Calibration tests cover the case where actualDays is far below declared estimate days and assert bias is `over`. | estado skipped no comprobado |
+| OO-212 | No cumple | skipped | Existing under-estimate coverage remains intact for actualDays above estimates. | estado skipped no comprobado |
+| OO-212 | No cumple | skipped | Lesson metadata and recommendation text match the bias direction. | estado skipped no comprobado |
+| OO-212 | No cumple | skipped | The test uses deterministic timestamps and does not rely on wall-clock race conditions. | estado skipped no comprobado |
+| OO-211 | No cumple | gap | Release playbook includes a mandatory `Deployment Risk Challenge` section. | gap de automatizacion/evidencia |
+| OO-211 | No cumple | planned | Release must flag deployment smells such as excessive rollout waves for story size, rollback complexity disproportionate to change size, production-impacting config/API changes, security boundary changes, and missing observability. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-211 | No cumple | gap | If risk smells exceed a configurable threshold, Release records a blocking review or action-policy approval request instead of rubber-stamping the runbook. | gap de automatizacion/evidencia |
+| OO-211 | No cumple | gap | Release readiness gate surfaces unresolved deployment challenges. | gap de automatizacion/evidencia |
+| OO-211 | No cumple | planned | Existing release behavior remains unchanged when no challenge findings are present. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-210 | No cumple | gap | `orchestra workflow run` supports `--mode new\|redo\|sign-off`, defaulting to `new`. | gap de automatizacion/evidencia |
+| OO-210 | No cumple | gap | `redo` mode prepends a clear rework protocol to the active phase prompt. | gap de automatizacion/evidencia |
+| OO-210 | No cumple | gap | In `redo` mode, decisions tagged `OBSERVATION:` are treated as mandatory review inputs. | gap de automatizacion/evidencia |
+| OO-210 | No cumple | gap | Phase output must address every matching observation with quote, accept/reject/defer, reasoning, and evidence. | gap de automatizacion/evidencia |
+| OO-210 | No cumple | gap | `sign-off` mode must not silently clear unresolved observations. | gap de automatizacion/evidencia |
+| OO-210 | No cumple | gap | Tests cover rerunning from Architect with observation decisions and assert the prompt/output contract includes each observation. | gap de automatizacion/evidencia |
+| OO-209 | No cumple | skipped | Developer playbook requires an `Architectural Concerns (inherited)` section for inherited design drift. | estado skipped no comprobado |
+| OO-209 | No cumple | skipped | Developer playbook requires an `Architectural Concerns (self-imposed)` section for new abstractions, files, metadata, APIs, config, scripts, or workflow changes proposed by Developer. | estado skipped no comprobado |
+| OO-209 | No cumple | skipped | Each self-imposed item must justify why existing project patterns or simpler alternatives are insufficient. | estado skipped no comprobado |
+| OO-209 | No cumple | skipped | Empty concern sections must be explicit rather than omitted. | estado skipped no comprobado |
+| OO-209 | No cumple | skipped | Phase structured output can carry challenge findings forward to QA and Release. | estado skipped no comprobado |
+| OO-209 | No cumple | skipped | Tests verify Developer prompt/playbook includes both inherited and self-imposed review requirements. | estado skipped no comprobado |
+| OO-224 | No cumple | planned | Recommendations explain rationale; frontend or UX-risk tasks include UX/accessibility validation; docs/site/release-risk tasks include technical writer or release-manager validation; manual workflow.phaseSequence remains authoritative; CLI exposes dry-run and JSON phase-plan output; tests cover default, frontend, docs/release, duplicate prevention, and manual override behavior | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Given project signals from `orchestra detect`, the planner recommends additional phases only when the task scope or risk justifies them. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Given a frontend or UX-risk task, the recommendation includes UX/accessibility review before QA or release. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Given docs/site/release-impacting work, the recommendation includes technical writer or release-manager validation where appropriate. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Manual `workflow.phaseSequence` remains authoritative and is not overwritten silently. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | The CLI exposes a dry-run/JSON view of the recommended phase plan and rationale. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Unit tests cover default behavior, frontend risk, docs/release risk, duplicate phase prevention, and manual override behavior. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Given project signals from `orchestra detect`, the planner recommends additional phases only when the task scope or risk justifies them. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Given a frontend or UX-risk task, the recommendation includes UX/accessibility review before QA or release. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Given docs/site/release-impacting work, the recommendation includes technical writer or release-manager validation where appropriate. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Manual `workflow.phaseSequence` remains authoritative and is not overwritten silently. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | The CLI exposes a dry-run/JSON view of the recommended phase plan and rationale. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-224 | No cumple | planned | Unit tests cover default behavior, frontend risk, docs/release risk, duplicate phase prevention, and manual override behavior. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-225 | No cumple | planned | Initialization creates focused playbooks or sections when relevant; user-edited playbooks are preserved unless an explicit force or refresh path is used; provider-backed phase prompts load only active playbook content; workflow render and runtime brief show loaded playbook source; tests cover init, idempotency, missing fallback, and prompt inclusion boundaries | solo planificado; sin evidencia ejecutada enlazada |
+| OO-225 | No cumple | gap | Default initialization creates focused playbooks or playbook sections for `qa-checklist`, `regression`, `smoke-test`, `doc-review`, `release-promote`, and `ux-review` when those phases/roles are active. | gap de automatizacion/evidencia |
+| OO-225 | No cumple | planned | Existing user-edited playbooks are not overwritten unless an explicit force/refresh path is used. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-225 | No cumple | planned | Provider-backed phase prompts load only the active phase/specialized playbook content, not every playbook. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-225 | No cumple | planned | Workflow render/runtime brief output shows which playbook source was loaded. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-225 | No cumple | planned | Tests cover initialization, idempotency, missing playbook fallback, and prompt inclusion boundaries. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-226 | No cumple | planned | Refresh command or mode updates managed runtime bootstrap files, default playbooks, instruction manifests, and generated metadata from .agent-workflow/config.json; supports check dry-run force and json semantics; preserves user-authored content outside managed blocks; reports drift clearly and requires force before replacement; tests cover no-op, dry-run, check, force regeneration, path safety, and user content preservation | solo planificado; sin evidencia ejecutada enlazada |
+| OO-226 | No cumple | planned | A command or documented mode refreshes managed runtime bootstrap files, default playbooks, instruction manifests, and generated metadata from existing `.agent-workflow/config.json`. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-226 | No cumple | planned | The command supports `--check`, `--dry-run`, `--force`, and `--json` semantics consistent with existing instruction commands. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-226 | No cumple | planned | User-authored content outside managed blocks is preserved. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-226 | No cumple | planned | Drifted managed blocks are reported clearly, with force required before replacement. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-226 | No cumple | planned | Tests cover no-op refresh, dry-run/check output, force regeneration, path safety, and preservation of user content. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-227 | No cumple | planned | Non-JSON workflow runs emit phase start provider model retry fallback elapsed heartbeat and completion or failure messages; JSON output remains clean; timeout messages include phase id provider model elapsed time and sanitized failure reason; Web/API status exposes active phase progress data; tests cover heartbeat, JSON silence, fallback diagnostics, and timeout messaging | solo planificado; sin evidencia ejecutada enlazada |
+| OO-227 | No cumple | planned | Non-JSON workflow runs emit phase start, provider/model attempt, retry/fallback, elapsed-time heartbeat, and phase completion/failure messages. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-227 | No cumple | planned | JSON output remains machine-readable and does not interleave heartbeat text into JSON payloads. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-227 | No cumple | planned | Timeout messages include phase id, provider/model, elapsed time, and sanitized failure reason. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-227 | No cumple | planned | Web/API run status can expose enough state for the UI to show active phase progress. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-227 | No cumple | planned | Tests cover non-JSON heartbeat behavior, JSON silence, provider fallback diagnostics, and timeout messaging. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-228 | No cumple | planned | Cursor-scoped command or skill reports canvas sync status without requiring Cursor in non-Cursor workspaces; sync supports dry-run status and clean behavior with clear source and destination paths; paths are validated and never write outside workspace; non-Cursor projects get unsupported or no-op result; tests cover status dry-run sync clean path safety and non-Cursor fallback | solo planificado; sin evidencia ejecutada enlazada |
+| OO-228 | No cumple | planned | A Cursor-scoped command or skill can report canvas sync status without requiring Cursor in non-Cursor workspaces. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-228 | No cumple | planned | Sync supports dry-run/status/clean behavior and clear paths for cache source and repo destination. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-228 | No cumple | planned | The command validates paths and never writes outside the workspace. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-228 | No cumple | planned | Non-Cursor projects get a clear unsupported/no-op result. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-228 | No cumple | planned | Tests cover status, dry-run sync, clean, path safety, and non-Cursor fallback. | solo planificado; sin evidencia ejecutada enlazada |
+| OO-238 | No cumple | planned | Tests cover spawn limit, depth limit, queue/reject behavior, crash/timeout recovery, stale claim release, and context budget enforcement. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-228 | No cumple | planned | Tests cover prompt coverage | solo planificado; sin evidencia ejecutada enlazada |
+| GH-228 | No cumple | planned | stale entries | solo planificado; sin evidencia ejecutada enlazada |
+| GH-228 | No cumple | planned | lesson promotion | solo planificado; sin evidencia ejecutada enlazada |
+| GH-228 | No cumple | planned | and task context retrieval. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-228 | No cumple | planned | Prompt registry entries link to tasks issues decisions evidence reviews and changed paths; validation reports missing or stale prompt coverage with actionable remediation; lesson promotion can reference originating task and artifact; memory context retrieval prefers relevant entries without prompt bloat; tests cover prompt coverage stale entries lesson promotion and task context retrieval | solo planificado; sin evidencia ejecutada enlazada |
+| GH-229 | No cumple | planned | Unit and workflow tests cover success | solo planificado; sin evidencia ejecutada enlazada |
+| GH-229 | No cumple | planned | retry | solo planificado; sin evidencia ejecutada enlazada |
+| GH-229 | Cumple | covered | fallback | evidencia=1 |
+| GH-229 | No cumple | planned | timeout | solo planificado; sin evidencia ejecutada enlazada |
+| GH-229 | No cumple | planned | denied permission | solo planificado; sin evidencia ejecutada enlazada |
+| GH-229 | No cumple | planned | and blocked fallback cases. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-229 | No cumple | planned | Runtime adapter contracts define retry fallback timeout permission and failure semantics; provider failures are sanitized and recorded consistently across CLI web API and evidence; timeout handling distinguishes provider timeout phase timeout and run timeout; permission modes are testable without invoking real external providers; tests cover success retry fallback timeout denied permission and blocked fallback cases | solo planificado; sin evidencia ejecutada enlazada |
+| GH-230 | No cumple | planned | Tests cover browser | solo planificado; sin evidencia ejecutada enlazada |
+| GH-230 | Cumple | covered | CLI | evidencia=1 |
+| GH-230 | Cumple | covered | shell | evidencia=1 |
+| GH-230 | No cumple | planned | and non-applicable task scenarios. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-230 | No cumple | planned | QA phase selects Playwright CLI shell or other E2E style based on task scope and project signals; generated E2E artifacts are workflow-local and executable; QA handoff includes commands setup assumptions and evidence references; non-applicable tasks record why E2E was skipped; tests cover browser CLI shell and non-applicable scenarios | solo planificado; sin evidencia ejecutada enlazada |
+| GH-231 | No cumple | planned | Tests cover unchanged | solo planificado; sin evidencia ejecutada enlazada |
+| GH-231 | No cumple | planned | drifted | solo planificado; sin evidencia ejecutada enlazada |
+| GH-231 | No cumple | planned | blocked | solo planificado; sin evidencia ejecutada enlazada |
+| GH-231 | No cumple | planned | forced | solo planificado; sin evidencia ejecutada enlazada |
+| GH-231 | No cumple | planned | missing | solo planificado; sin evidencia ejecutada enlazada |
+| GH-231 | No cumple | planned | and unsupported target cases. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-231 | No cumple | planned | Refresh reports include per-file before and after status managed block identity target runtime and recommended command; check mode exits with machine-readable drift status; dry-run shows planned writes without touching files; force mode preserves user-authored content outside managed blocks; tests cover unchanged drifted blocked forced missing and unsupported target cases | solo planificado; sin evidencia ejecutada enlazada |
+| GH-232 | No cumple | planned | Tests cover rendering active | solo planificado; sin evidencia ejecutada enlazada |
+| GH-232 | No cumple | planned | failed | solo planificado; sin evidencia ejecutada enlazada |
+| GH-232 | No cumple | planned | paused | solo planificado; sin evidencia ejecutada enlazada |
+| GH-232 | No cumple | planned | and completed workflow runs. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-232 | No cumple | planned | Active runs show current phase role provider model elapsed time and latest status; failed provider phases show sanitized reason and fallback state; paused gates and resumable phases remain visually distinct from running phases; UI remains responsive on mobile and desktop; tests cover active failed paused and completed workflow runs | solo planificado; sin evidencia ejecutada enlazada |
+| GH-233 | No cumple | planned | Tests cover CLI available | solo planificado; sin evidencia ejecutada enlazada |
+| GH-233 | No cumple | planned | CLI missing with MCP fallback | solo planificado; sin evidencia ejecutada enlazada |
+| GH-233 | No cumple | planned | unsupported tracker | solo planificado; sin evidencia ejecutada enlazada |
+| GH-233 | No cumple | planned | and failed fallback. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-233 | No cumple | planned | Tracking integration contract supports CLI transport and MCP-skill fallback transport; GitHub remains supported while Jira GitLab and Bitbucket can be added without changing core workflow logic; missing CLI but available MCP skill produces clear fallback path; sync records selected transport in evidence or events; tests cover CLI available CLI missing with MCP fallback unsupported tracker and failed fallback | solo planificado; sin evidencia ejecutada enlazada |
+| GH-234 | No cumple | planned | Tests cover ready | solo planificado; sin evidencia ejecutada enlazada |
+| GH-234 | No cumple | planned | blocked | solo planificado; sin evidencia ejecutada enlazada |
+| GH-234 | No cumple | planned | accepted-risk | solo planificado; sin evidencia ejecutada enlazada |
+| GH-234 | No cumple | planned | and rollback-required release paths. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-234 | No cumple | planned | Release readiness summarizes acceptance criteria tests security checks docs migration risk rollback and known gaps; release candidate artifacts can be generated and reviewed from CLI and web console; required evidence gates can be configured by project risk profile; release notes and changelog inputs are generated from tasks commits and evidence; tests cover ready blocked accepted-risk and rollback-required paths | solo planificado; sin evidencia ejecutada enlazada |
+| GH-235 | No cumple | planned | Documentation validation passes. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-235 | No cumple | planned | Docs cover workflow phase plans specialized playbooks generated file refresh provider progress and Cursor canvas sync; public site includes concise examples and command references; docs state CLI-only web-supported and runtime-specific behavior; screenshots or examples are updated where web console behavior changed; documentation validation passes | solo planificado; sin evidencia ejecutada enlazada |
+| GH-236 | No cumple | planned | Tests cover automatic sizing | solo planificado; sin evidencia ejecutada enlazada |
+| GH-236 | No cumple | planned | manual override | solo planificado; sin evidencia ejecutada enlazada |
+| GH-236 | Cumple | covered | missing sizing | evidencia=1 |
+| GH-236 | No cumple | planned | and benchmark integration. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-236 | No cumple | planned | Architect phase can propose T-shirt size with rationale; developer phase can propose or confirm story points with implementation risk notes; estimates are stored as structured events and feed benchmark and burndown reports; calibration compares predicted vs actual effort and records lessons when bias repeats; tests cover automatic sizing manual override missing sizing and benchmark integration | solo planificado; sin evidencia ejecutada enlazada |
+| GH-237 | No cumple | gap | Docs include setup examples for local and API-backed providers. | gap de automatizacion/evidencia |
+| GH-237 | No cumple | gap | Each supported runtime has documented configuration capability detection invocation plan and permission model; runtime selection is provider-agnostic and can mix parent execution with delegated providers; missing runtime dependencies produce clear remediation messages; deterministic adapter tests cover each runtime path; docs include setup examples for local and API-backed providers | gap de automatizacion/evidencia |
+| GH-292 | No cumple | planned | npm test passes | solo planificado; sin evidencia ejecutada enlazada |
+| GH-293 | No cumple | planned | tests prove placeholder provider output does not silently pass as done | solo planificado; sin evidencia ejecutada enlazada |
+| RUN-VS-WORKFLOW | No cumple | planned | tests cover the updated help copy | solo planificado; sin evidencia ejecutada enlazada |
+| GH-294 | No cumple | planned | npm run lint passes; npm run precommit passes; MCP redacted errors preserve cause/context; Mermaid regex lint issues are removed without behavior regression | solo planificado; sin evidencia ejecutada enlazada |
+| GH-294 | No cumple | planned | npm run lint passes with zero errors. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-294 | No cumple | planned | npm run precommit passes locally. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-294 | No cumple | planned | src/tool-commands.ts preserves caught error cause or equivalent context when redacting MCP errors. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-294 | No cumple | planned | web-console/src/main.jsx regex lint issues are removed without changing Mermaid parsing behavior. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-294 | No cumple | planned | Add or update focused tests only if behavior changes. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-298 | No cumple | planned | React console is split into focused modules; dashboard supports section-level endpoint failure states; task/evidence/workflow/provider/release/recovery flows remain available; responsive/accessibility behavior remains covered; E2E proves one partial API failure does not blank the console | solo planificado; sin evidencia ejecutada enlazada |
+| GH-298 | No cumple | planned | Split web-console/src/main.jsx into focused components or feature modules. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-298 | No cumple | gap | Dashboard loading handles partial endpoint failures with section-level empty/error states. | gap de automatizacion/evidencia |
+| GH-298 | No cumple | planned | Existing task, evidence, workflow, provider, release, and recovery flows remain available. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-298 | No cumple | planned | Responsive and accessibility behavior remains covered. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-298 | No cumple | gap | E2E coverage proves at least one partial API failure does not blank the whole console. | gap de automatizacion/evidencia |
+| GH-297 | No cumple | planned | Architecture decision records module boundaries; at least one high-risk oversized module is split into focused services; public command behavior is preserved; focused tests cover moved behavior; unrelated churn is avoided | solo planificado; sin evidencia ejecutada enlazada |
+| GH-297 | No cumple | planned | Identify stable module boundaries before editing and record the architecture decision. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-297 | No cumple | planned | Extract at least one high-risk oversized module into focused services without changing public command behavior. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-297 | No cumple | planned | Preserve barrel/public imports and avoid deep import churn where possible. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-297 | No cumple | planned | Focused tests continue to cover moved behavior. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-297 | No cumple | planned | No unrelated refactor or formatting churn. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-295 | No cumple | gap | Release check vs direct npm test discrepancy is understood and fixed; version/tag mismatch is documented or resolved; smoke and rollback evidence guidance is clear; historical uncovered AC output has candidate/task scoping or actionable remediation | gap de automatizacion/evidencia |
+| GH-295 | No cumple | planned | Investigate and fix the release check vs direct npm test discrepancy. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-295 | No cumple | planned | Release check reports command failure details compactly and actionably. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-295 | No cumple | planned | Version/tag mismatch behavior is documented or resolved for the current release path. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-295 | No cumple | planned | Smoke and rollback evidence requirements are clear in CLI output and docs. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-295 | No cumple | planned | Release readiness does not drown users in historical uncovered AC without a clear candidate/task scope or remediation path. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-296 | No cumple | planned | Target first-run journey and success metric are defined; README/adoption/site copy provide shorter guided path; full governed production path remains documented; examples use real commands; docs/help tests cover the changed copy | solo planificado; sin evidencia ejecutada enlazada |
+| GH-296 | No cumple | planned | Define the target first-run journey and success metric. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-296 | No cumple | planned | Provide a shorter guided path or command copy that gets users from init to visible evidence with fewer concepts upfront. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-296 | No cumple | planned | Keep the full governed path documented for production work. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-296 | No cumple | planned | Update README, adoption guide, and public site copy consistently. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-296 | No cumple | planned | Add tests that validate help/docs examples use real commands. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-299 | No cumple | planned | task update --qa-plan and --qa-execution-status persist qaGate metadata; command manifest/help lists QA gate metadata flags; qa-release gate passes after evidence plus task update metadata without manual file edits; regression tests cover passed and deferred QA execution metadata | solo planificado; sin evidencia ejecutada enlazada |
+| GH-300 | No cumple | planned | Audit identifies duplicated dedupe/normalization patterns with affected files and chosen consolidation boundaries | solo planificado; sin evidencia ejecutada enlazada |
+| GH-300 | No cumple | planned | Existing public behavior is covered by regression tests or focused existing tests | solo planificado; sin evidencia ejecutada enlazada |
+| GH-300 | No cumple | planned | Shared helpers are introduced only where at least two modules benefit and behavior stays identical | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-005 | No cumple | planned | Filters expose state owner text and internal workflow visibility | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-007 | No cumple | planned | Mobile layout remains readable without horizontal overflow | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-007 | No cumple | planned | Recovery Center Decisions and Release Readiness do not create awkward cross-panel misalignment | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-007 | No cumple | planned | Work Quality Release Operations and Settings content are separated into scan-friendly sections | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-007 | No cumple | planned | Work Release Evidence Operations and Settings content are separated into scan-friendly sections | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-008 | No cumple | planned | Mobile remains single-column with no horizontal overflow | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-009 | No cumple | planned | Mobile remains single-column without horizontal overflow | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-010 | No cumple | planned | Roles does not dominate the first Settings viewport | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-011 | No cumple | planned | Release summary markdown renders as structured readable content without unsafe HTML | solo planificado; sin evidencia ejecutada enlazada |
+| UX-UI-012 | No cumple | planned | Existing Playwright flows remain usable through the new navigation | solo planificado; sin evidencia ejecutada enlazada |
+| ESTIMATE-BENCHMARK-CALIBRATION | No cumple | skipped | estimate reads completed workflow benchmark data when at least two comparable runs exist. | estado skipped no comprobado |
+| ESTIMATE-BENCHMARK-CALIBRATION | No cumple | skipped | output includes baseline estimate, historical median, variance, and calibration note. | estado skipped no comprobado |
+| ESTIMATE-BENCHMARK-CALIBRATION | No cumple | skipped | --ignore-history bypasses calibration. | estado skipped no comprobado |
+| ESTIMATE-BENCHMARK-CALIBRATION | No cumple | skipped | --json includes historicalMedian, variance, and calibrationNote fields. | estado skipped no comprobado |
+| ESTIMATE-BENCHMARK-CALIBRATION | No cumple | skipped | tests cover no history, one run, two-plus runs, and ignore-history. | estado skipped no comprobado |
+| ESTIMATE-BENCHMARK-CALIBRATION | No cumple | skipped | Backlog Item ID: ESTIMATE-BENCHMARK-CALIBRATION | estado skipped no comprobado |
+| DEV-STANDARDS-GENERALIZE | No cumple | skipped | Open Orchestra rules document stack-agnostic versions of the reusable developer standards. | estado skipped no comprobado |
+| DEV-STANDARDS-GENERALIZE | No cumple | skipped | Examples cover Java, .NET, TypeScript, and Python without importing Apex-only terminology. | estado skipped no comprobado |
+| DEV-STANDARDS-GENERALIZE | No cumple | skipped | Apex/Salesforce-specific items remain out of Open Orchestra core. | estado skipped no comprobado |
+| DEV-STANDARDS-GENERALIZE | No cumple | skipped | docs/setup-agents-applicability-review.md is used as the source review artifact. | estado skipped no comprobado |
+| DEV-STANDARDS-GENERALIZE | No cumple | skipped | Backlog Item ID: DEV-STANDARDS-GENERALIZE | estado skipped no comprobado |
+| VISUAL-VALIDATION-GATE | No cumple | planned | [ ] Orchestrator supports a `visual_validation` gate type | solo planificado; sin evidencia ejecutada enlazada |
+| VISUAL-VALIDATION-GATE | No cumple | planned | [ ] Gate is triggered automatically when sub-agent manifest includes visual tool writes | solo planificado; sin evidencia ejecutada enlazada |
+| VISUAL-VALIDATION-GATE | No cumple | planned | [ ] Assertion suite is configurable per project (YAML or JSON) | solo planificado; sin evidencia ejecutada enlazada |
+| VISUAL-VALIDATION-GATE | No cumple | planned | [ ] Auto-remediation runs for fixable issues (duplicates, overlaps, fallback text) | solo planificado; sin evidencia ejecutada enlazada |
+| VISUAL-VALIDATION-GATE | No cumple | planned | [ ] PNG export artifact is attached to workflow evidence | solo planificado; sin evidencia ejecutada enlazada |
+| VISUAL-VALIDATION-GATE | No cumple | planned | [ ] Gate blocks phase completion if non-remediable issues are found | solo planificado; sin evidencia ejecutada enlazada |
+| VISUAL-VALIDATION-GATE | No cumple | planned | [ ] Works with Lucid MCP; extensible to Figma, draw.io, Mermaid | solo planificado; sin evidencia ejecutada enlazada |
+| VISUAL-VALIDATION-GATE | No cumple | planned | Gate exists and is discoverable; CLI validates visual manifests; Lucid rotations default to radians; JSON config supports thresholds; PNG export evidence is recorded; gate blocks missing/failed validation; tests cover passing/blocking paths. | solo planificado; sin evidencia ejecutada enlazada |
+| SETUP-IMPORT-ROLE-ALIASES | No cumple | skipped | setup-agents import maps pm, product, product_manager aliases to product_manager/product_owner as appropriate. | estado skipped no comprobado |
+| SETUP-IMPORT-ROLE-ALIASES | No cumple | skipped | Existing setup-agents import tests cover PM and PO aliases. | estado skipped no comprobado |
+| SETUP-IMPORT-ROLE-ALIASES | No cumple | skipped | Imported tasks do not persist non-canonical ownerRole values for PM/PO. | estado skipped no comprobado |
+| SETUP-IMPORT-ROLE-ALIASES | No cumple | skipped | Backlog Item ID: SETUP-IMPORT-ROLE-ALIASES | estado skipped no comprobado |
+| RULES-PLAYBOOKS-HEALTH | No cumple | planned | health or status output includes generated guidance health. | solo planificado; sin evidencia ejecutada enlazada |
+| RULES-PLAYBOOKS-HEALTH | No cumple | planned | output identifies stale managed blocks or generated files with actionable refresh command. | solo planificado; sin evidencia ejecutada enlazada |
+| RULES-PLAYBOOKS-HEALTH | No cumple | planned | output identifies missing expected playbooks and suggests scaffolding or authoring next step. | solo planificado; sin evidencia ejecutada enlazada |
+| RULES-PLAYBOOKS-HEALTH | No cumple | planned | --json exposes structured health data. | solo planificado; sin evidencia ejecutada enlazada |
+| RULES-PLAYBOOKS-HEALTH | No cumple | planned | Backlog Item ID: RULES-PLAYBOOKS-HEALTH | solo planificado; sin evidencia ejecutada enlazada |
+| WORKFLOW-DOCS-PLAYBOOKS-GATES | No cumple | planned | docs include an end-to-end Open Orchestra workflow narrative from init to release evidence. | solo planificado; sin evidencia ejecutada enlazada |
+| WORKFLOW-DOCS-PLAYBOOKS-GATES | No cumple | planned | docs include phase matrix with roles, gates, and human approval points. | solo planificado; sin evidencia ejecutada enlazada |
+| WORKFLOW-DOCS-PLAYBOOKS-GATES | No cumple | planned | docs explain gate vs clarify with examples and decision table. | solo planificado; sin evidencia ejecutada enlazada |
+| WORKFLOW-DOCS-PLAYBOOKS-GATES | No cumple | planned | docs explain playbook authoring/resolution and link from command reference. | solo planificado; sin evidencia ejecutada enlazada |
+| WORKFLOW-DOCS-PLAYBOOKS-GATES | No cumple | planned | Backlog Item ID: WORKFLOW-DOCS-PLAYBOOKS-GATES | solo planificado; sin evidencia ejecutada enlazada |
+| PLAYBOOK-SCAFFOLD | No cumple | planned | command scaffolds missing workflow playbook stubs without overwriting existing files. | solo planificado; sin evidencia ejecutada enlazada |
+| PLAYBOOK-SCAFFOLD | No cumple | planned | supports phase filter and dry-run. | solo planificado; sin evidencia ejecutada enlazada |
+| PLAYBOOK-SCAFFOLD | No cumple | planned | stubs include phase role context and placeholder instruction sections. | solo planificado; sin evidencia ejecutada enlazada |
+| PLAYBOOK-SCAFFOLD | No cumple | planned | output lists created or skipped paths. | solo planificado; sin evidencia ejecutada enlazada |
+| PLAYBOOK-SCAFFOLD | No cumple | planned | tests cover all missing, skip existing, dry-run, and phase filter. | solo planificado; sin evidencia ejecutada enlazada |
+| PLAYBOOK-SCAFFOLD | No cumple | planned | Backlog Item ID: PLAYBOOK-SCAFFOLD | solo planificado; sin evidencia ejecutada enlazada |
+| QA-PLAN-READY-GATE | No cumple | planned | Ready-for-dev or execution is blocked when qaGate.plan is missing unless QA is explicitly not applicable or deferred with rationale and owner; QA plan covers acceptance criteria regression edge cases data automation and evidence; existing qa-release gate remains compatible; CLI and web messaging make the missing plan actionable; tests cover plan present missing and not-applicable paths. | solo planificado; sin evidencia ejecutada enlazada |
+| QA-AUTOMATION-ASSERTIONS | No cumple | planned | Automation evidence requires assertions over observable outcomes; web API CLI and integration categories define expected observable checks; external validation can be sandbox mock contract webhook event log or deferred with owner rationale; CLI evidence validates outputs files events and state not only exit code; QA evidence records gaps mocked boundaries and follow-up owner; existing evidence workflows remain compatible. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-331 | No cumple | planned | Workflow phases run and record mandatory memory hooks before implementation, handoff, QA, and release-significant phases; Handoffs include lessons consulted/applied and prompt registry entries consulted/updated; Final deliverable gates block substantial changed artifacts without prompt registry coverage or accepted rationale; Recoverable failures with verified fixes create a lesson learned automatically or require documented not-applicable rationale before closure; Model provenance records link to prompt registry entries or privacy-preserving prompt summary/rationale; CLI output and web UI make missing memory consumption actionable; Tests cover lesson read, lesson captured, prompt registry read/update required, missing consumption blocked, and recorded evidence passing. | solo planificado; sin evidencia ejecutada enlazada |
+| GH-332 | Cumple | covered | Every AC has evidence or deferred owner/rationale | evidencia=1 |
+| GH-332 | Cumple | covered | automation evidence validates observable outcomes | evidencia=1 |
+| GH-332 | Cumple | covered | evidence distinguishes AC coverage from smoke/regression | evidencia=1 |
+| GH-332 | Cumple | covered | standards and tests document the rule | evidencia=1 |
+| E2E-AC-RESULTS-HARDENING | No cumple | planned | CLI E2E captures stdout stderr and exit code | solo planificado; sin evidencia ejecutada enlazada |
+| E2E-AC-RESULTS-HARDENING | No cumple | planned | CLI E2E maps each scenario to acceptance criteria outcomes | solo planificado; sin evidencia ejecutada enlazada |
+| E2E-AC-RESULTS-HARDENING | No cumple | planned | web E2E status is documented against observable UI and API outcomes | solo planificado; sin evidencia ejecutada enlazada |
+| E2E-AC-RESULTS-HARDENING | No cumple | planned | focused tests pass | solo planificado; sin evidencia ejecutada enlazada |
+| ARCH-SALESFORCE-INTEGRATION-DIAGRAM | No cumple | skipped | Architecture diagram includes all requested systems and visible relationships | estado skipped no comprobado |
+| ARCH-SALESFORCE-INTEGRATION-DIAGRAM | No cumple | skipped | connectors | estado skipped no comprobado |
+| ARCH-SALESFORCE-INTEGRATION-DIAGRAM | No cumple | skipped | boundaries | estado skipped no comprobado |
+| ARCH-SALESFORCE-INTEGRATION-DIAGRAM | No cumple | skipped | and integration direction. | estado skipped no comprobado |

package/docs/runtime-llm-flow.md

@@ -42,6 +42,14 @@ workflow instructions for a specific runtime.
 commands and which task loop to run without copying the full manual into the
 always-loaded context.
 
+Open Orchestra does not own the entire instruction file. It appends or refreshes
+only the marked managed block between `<!-- open-orchestra:start ... -->` and
+`<!-- open-orchestra:end ... -->`. Project-authored rules before or after that
+block remain owned by the project and are preserved. If the managed block was
+edited manually, refresh reports drift and refuses to overwrite it unless
+`--force` is supplied; `--force` replaces only the managed block, not the rest of
+the file.
+
 Use explicit init targets when a project should generate runtime-specific files:
 
 ```bash