@openlife/cli 1.7.3

package/docs/RELEASE_ORGANIZATION_PLAN.md

@@ -0,0 +1,164 @@
+# OpenLife — Plano de Organização, Commits Atômicos e Release
+
+## Objetivo
+Organizar as mudanças atuais em lotes atômicos, com validação incremental e smoke test final.
+
+## Estratégia de Branch
+- Branch recomendada: `feat/reversa-designmd-full-openlife`
+- Regra: 1 commit por bloco funcional coeso.
+
+## Lote 1 — Core dual-mode + governança
+**Escopo**
+- `src/orchestrator/ExecutionIntent.ts`
+- `src/orchestrator/ExecutionRouter.ts`
+- `src/orchestrator/MissionState.ts`
+- `src/orchestrator/OrchestrationLoop.ts`
+- `src/orchestrator/Gatekeeper.ts`
+- `src/orchestrator/GovernanceLayer.ts`
+- `src/test_dual_mode.ts`
+- `src/test_service_guardrails_delete.ts`
+- `src/test_telegram_delete_guardrail.ts`
+
+**Commit**
+- `feat(core): dual-mode task/service + destructive guardrails`
+
+**Validação**
+- `npm run build`
+- `node dist/test_dual_mode.js`
+- `node dist/test_service_guardrails_delete.js`
+- `node dist/test_telegram_delete_guardrail.js`
+
+---
+
+## Lote 2 — Lifecycle de jobs + runtime registry
+**Escopo**
+- `src/orchestrator/JobLifecycle.ts`
+- `src/orchestrator/RuntimeRegistry.ts`
+- `src/test_job_lifecycle.ts`
+- `src/test_runtime_registry.ts`
+
+**Commit**
+- `feat(runtime): job lifecycle + runtime registry baseline`
+
+**Validação**
+- `npm run build`
+- `node dist/test_job_lifecycle.js`
+- `node dist/test_runtime_registry.js`
+
+---
+
+## Lote 3 — Service state, CLI surface e operações
+**Escopo**
+- `src/orchestrator/ServiceState.ts`
+- `src/index.ts` (comandos service/task/job/runtime)
+- `src/test_service_command_surface.ts`
+- `src/test_cli_service_commands.ts`
+
+**Commit**
+- `feat(cli): service/task/job/runtime command surface`
+
+**Validação**
+- `npm run build`
+- `node dist/test_service_command_surface.js`
+- `node dist/test_cli_service_commands.js`
+
+---
+
+## Lote 4 — Teammates + learning loop
+**Escopo**
+- `src/orchestrator/TeammateBoard.ts`
+- `src/orchestrator/SkillLearningLoop.ts`
+- `src/test_teammate_learning.ts`
+
+**Commit**
+- `feat(ops): teammate board + governed learning loop baseline`
+
+**Validação**
+- `npm run build`
+- `node dist/test_teammate_learning.js`
+
+---
+
+## Lote 5 — Reversa full + contracts
+**Escopo**
+- `src/reversa/ReversaExecutors.ts`
+- `src/reversa/ReversaContracts.ts`
+- `src/reversa/ReversaAgent.ts`
+- `src/test_reversa_lite.ts`
+- `src/test_reversa_full_execution.ts`
+- `src/test_reversa_contracts_e2e.ts`
+
+**Commit**
+- `feat(reversa): full phase executors + contract validation + final contracts`
+
+**Validação**
+- `npm run build`
+- `node dist/test_reversa_lite.js`
+- `node dist/test_reversa_full_execution.js`
+- `node dist/test_reversa_contracts_e2e.js`
+
+---
+
+## Lote 6 — DesignMD import/registry/mode + AIOBUILDER strict
+**Escopo**
+- `vendor/awesome-design-md/**`
+- `src/design/DesignMdImporter.ts`
+- `src/design/DesignMdRegistry.ts`
+- `src/design/DesignMdMode.ts`
+- `src/index.ts` (comandos designmd/reversa export-json/aiobuilder strict)
+- `src/test_designmd_import_registry.ts`
+- `src/test_designmd_mode.ts`
+- `src/test_designmd_mode_workspace.ts`
+- `src/test_reversa_export_and_strict.ts`
+
+**Commit**
+- `feat(designmd): vendor import + registry/mode + aiobuilder strict contracts`
+
+**Validação**
+- `npm run build`
+- `node dist/test_designmd_import_registry.js`
+- `node dist/test_designmd_mode.js`
+- `node dist/test_designmd_mode_workspace.js`
+- `node dist/test_reversa_export_and_strict.js`
+
+---
+
+## Lote 7 — Documentação
+**Escopo**
+- `docs/CHANGELOG_FEATURE_ROLLOUT_DESIGNMD.md`
+- `docs/OPENLIFE_DUAL_MODE_IMPLEMENTATION_PLAN.md`
+- `docs/PAPERCLIP_OPENLIFE_INVESTIGATION.md`
+- `docs/RELEASE_ORGANIZATION_PLAN.md`
+
+**Commit**
+- `docs: rollout changelog + dual-mode/reversa/designmd plans`
+
+---
+
+## Smoke test final (único comando)
+```bash
+npm run build && \
+node dist/test_dual_mode.js && \
+node dist/test_service_guardrails_delete.js && \
+node dist/test_telegram_delete_guardrail.js && \
+node dist/test_job_lifecycle.js && \
+node dist/test_runtime_registry.js && \
+node dist/test_service_command_surface.js && \
+node dist/test_cli_service_commands.js && \
+node dist/test_teammate_learning.js && \
+node dist/test_reversa_lite.js && \
+node dist/test_reversa_full_execution.js && \
+node dist/test_reversa_contracts_e2e.js && \
+node dist/test_designmd_import_registry.js && \
+node dist/test_designmd_mode.js && \
+node dist/test_designmd_mode_workspace.js && \
+node dist/test_reversa_export_and_strict.js
+```
+
+## Checklist de release interno
+- [ ] Todos os lotes com build/testes verdes
+- [ ] `openlife reversa run-all --note "pre-release"` executado
+- [ ] `openlife reversa export-json` gerado
+- [ ] `openlife aiobuilder generate-ui dashboard --strict-contracts` OK
+- [ ] Revisão de artefatos em `.reversa/` e `.artifacts/workspaces/default/`
+- [ ] Aprovação final do Rafa antes de push/deploy

package/docs/audit/CLI-EXECUTION-RESULTS.md

@@ -0,0 +1,113 @@
+# CLI Execution Results — Phase 2
+
+**Generated:** 2026-05-11
+**Phase:** 2 — CLI End-to-End Exercise
+**Requirements covered:** CLI-01, CLI-02, CLI-03, CLI-04, CLI-05
+**Test files added:** `test_cli_help_surface.ts`, `test_cli_diagnostics.ts`, `test_cli_crud_roundtrip.ts`
+**Test suite:** 56 → 59 tests, all green via `npm run test:all`
+
+---
+
+## Summary
+
+| Category | Tested | PASS | GRACEFUL | GAP | FAIL→Story |
+|----------|--------|------|----------|-----|------------|
+| Help/Version (45 groups + 2 root) | 47 | 47 | 0 | 0 | 0 |
+| Diagnostics (read-only commands) | 22 | 12 | 6 | 4 | 2 KNOWN BUGS |
+| CRUD Roundtrips (create→list→cleanup) | 6 | 6 | 0 | 0 | 0 |
+| **Total in Phase 2** | **75** | **65** | **6** | **4** | **2** |
+
+---
+
+## CLI-01: Help/Version Surface ✓
+
+Test: `src/test_cli_help_surface.ts` — exits 0 in ~15s
+
+- Root `--help`: PASS (4,200+ bytes output)
+- Root `--version`: PASS (returns `1.0.0`)
+- All 45 top-level groups responded to `--help` with exit 0 and non-empty output
+
+**Slow `--help` invocations (>3s)** — bookmark for Phase 5 lazy-load review:
+
+| Group | Latency | Likely cause |
+|-------|---------|--------------|
+| `pilot --help` | 4175ms | Module-scope import of heavy enterprise classes |
+| `learning --help` | 3311ms | Possible eager import of learning loop subsystem |
+| `plugin --help` | 3249ms | Possible eager import of plugin manifest store |
+
+**Why this is a regression risk:** `CLAUDE.md` documents "lazy imports for heavy runtime classes" as a critical invariant. `--help` should be near-instant. These three suggest violation of the pattern.
+
+---
+
+## CLI-02 / CLI-03 / CLI-04: Diagnostics ✓ with 2 known bugs
+
+Test: `src/test_cli_diagnostics.ts` — exits 0 in ~34s
+
+### PASS (exit 0 required, all green)
+- `catalog list`
+- `agents list`, `squads list`, `skills list`
+- `models status`
+- `memory status`, `memory provider-list`
+- `governance status`, `governance audit`
+- `media route-status`
+- `orchestration contract`, `orchestration snapshot`
+
+### GRACEFUL (exit 0 or 1 accepted; doctors that degrade gracefully)
+- `catalog doctor` — may warn about test-* artifacts (P3 design issue, not regression)
+- `system doctor` — warns about optional CLIs (codex/gemini)
+- `system status`
+- `media doctor`
+- `enterprise` — aggregate doctor
+- `swarm consensus-test`
+
+### GAP — known broken, deferred (recorded for Phase 5)
+- `phase1-check` — **HANGS** for >35s, killed by SIGKILL. **P1 BUG-01**.
+- `mcp status` (without `--real`) — exits 1 with "Use --real para obter o status determinístico". UX/default behavior bug. **P2 BUG-02**.
+
+### GAP — requires external state (Phase 4)
+- `system doctor-world` — requires LLM call
+- `benchmark report` — requires prior `benchmark run` to have data
+
+---
+
+## CLI-05: CRUD Roundtrips ✓
+
+Test: `src/test_cli_crud_roundtrip.ts` — exits 0 in ~19s, **all 6 entity types create+list+cleanup correctly**
+
+| Entity | Create command | List command | Result |
+|--------|----------------|--------------|--------|
+| Agent | `agents create <id> --role specialist` | `agents list` | PASS, cleanup OK |
+| Squad | `squads create <id> --domain general` | `squads list` | PASS, cleanup OK |
+| Skill | `skills create <id>` | `skills list` | PASS, cleanup OK |
+| MCP | `mcp create <id> --transport stdio --entry echo` | `mcp status --real` | PASS, cleanup OK |
+| Team | `teams upsert <id> --domain general --sla balanced` | `teams list` | PASS, cleanup OK |
+| Network | `networks upsert <id> --domain general --capabilities cap1,cap2` | `networks list` | PASS, cleanup OK |
+
+**Cleanup hygiene:** Test uses unique IDs (`test-crud-${pid}-${timestamp}`) and removes each created entity. `.catalog/` is byte-clean after run.
+
+---
+
+## Phase 2 Findings (queued for Phase 5)
+
+See `.planning/phase-2/FINDINGS.md` for the full bug list with severity and reproduction steps.
+
+---
+
+## What's NOT covered in Phase 2 (intentional, by phase boundary)
+
+- Commands requiring live LLM (`ask`, `route intent`, `swarm run`, `governance risk-check`) → **Phase 4**
+- Commands requiring daemon (`start --daemon`, Telegram webhook flow) → **Phase 4**
+- Subsystem unit tests (Brain/Gateway internals beyond what's in `test_brain_*`, `test_daemon_sigterm`, `test_trigger_basic_auth`) → **Phase 3**
+- 105 DRIFT-HIGH undocumented commands → **Phase 5** (doc rewrites)
+
+---
+
+## Suite Health After Phase 2
+
+```
+Before: 56 tests, all green, ~4 min runtime
+After:  59 tests, all green, ~6 min runtime
+Delta:  +3 tests, +~2 min (most of the delta is cold-start cost of 73 subprocess spawns across the new tests)
+```
+
+Pre-existing pollution in `.artifacts/` and `.catalog/test-*` MUST be cleaned before `npm run test:all` runs cleanly. This is **P3 tech debt** (documented in CONCERNS.md), not a regression. A `pretest` hook that cleans these would fix it — recommended for Phase 5.