@kal-elsam/kairo-runtime 0.1.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +937 -0
- package/bin/harness.js +11 -0
- package/bin/kairo-runtime.js +11 -0
- package/bin/kairo.js +11 -0
- package/global-template/components/catalog.json +25 -0
- package/global-template/components/orchestrator/orchestrator.md +55 -0
- package/global-template/components/sdd-core/handoff.md +38 -0
- package/global-template/components/sdd-core/spec-sizing.md +38 -0
- package/global-template/components/sdd-core/workflow.md +31 -0
- package/global-template/core/orchestrator.md +55 -0
- package/package.json +64 -0
- package/prompts/HARNESS_ENFORCEMENT_UPGRADE.md +11 -0
- package/prompts/HARNESS_ENTERPRISE.md +33 -0
- package/prompts/HARNESS_INSTALLER_MASTER.md +853 -0
- package/prompts/HARNESS_MINIMAL.md +30 -0
- package/prompts/HARNESS_SPEC_SIZING_UPGRADE.md +41 -0
- package/prompts/HARNESS_STANDARD.md +35 -0
- package/prompts/HARNESS_UNIVERSAL_PARITY.md +56 -0
- package/repo-template/.claude/agents/adapter-parity-reviewer.md +17 -0
- package/repo-template/.claude/agents/architect.md +17 -0
- package/repo-template/.claude/agents/eval-engineer.md +17 -0
- package/repo-template/.claude/agents/reviewer.md +17 -0
- package/repo-template/.claude/agents/sdd-reviewer.md +17 -0
- package/repo-template/.claude/agents/sdd-spec-writer.md +17 -0
- package/repo-template/.claude/skills/checkpoint/SKILL.md +13 -0
- package/repo-template/.claude/skills/eval-design/SKILL.md +3 -0
- package/repo-template/.claude/skills/evals/SKILL.md +13 -0
- package/repo-template/.claude/skills/loop-engineering/SKILL.md +16 -0
- package/repo-template/.claude/skills/maintainability-review/SKILL.md +3 -0
- package/repo-template/.claude/skills/quality-gate-review/SKILL.md +3 -0
- package/repo-template/.claude/skills/sdd/SKILL.md +13 -0
- package/repo-template/.claude/skills/spec-complexity-classifier/SKILL.md +10 -0
- package/repo-template/.claude/skills/spec-escalation-review/SKILL.md +10 -0
- package/repo-template/.claude/skills/spec-intake/SKILL.md +10 -0
- package/repo-template/.claude/skills/tdd/SKILL.md +13 -0
- package/repo-template/.claude/skills/trust-review/SKILL.md +3 -0
- package/repo-template/.codex/skills/adapter-parity/SKILL.md +29 -0
- package/repo-template/.codex/skills/checkpoint/SKILL.md +36 -0
- package/repo-template/.codex/skills/eval-design/SKILL.md +3 -0
- package/repo-template/.codex/skills/evals/SKILL.md +28 -0
- package/repo-template/.codex/skills/loop-engineering/SKILL.md +24 -0
- package/repo-template/.codex/skills/maintainability-review/SKILL.md +3 -0
- package/repo-template/.codex/skills/quality-gate-review/SKILL.md +3 -0
- package/repo-template/.codex/skills/sdd/SKILL.md +30 -0
- package/repo-template/.codex/skills/spec-complexity-classifier/SKILL.md +10 -0
- package/repo-template/.codex/skills/spec-escalation-review/SKILL.md +10 -0
- package/repo-template/.codex/skills/spec-intake/SKILL.md +10 -0
- package/repo-template/.codex/skills/tdd/SKILL.md +27 -0
- package/repo-template/.codex/skills/trust-review/SKILL.md +3 -0
- package/repo-template/.cursor/agents/adapter-parity-reviewer.md +29 -0
- package/repo-template/.cursor/agents/architect.md +27 -0
- package/repo-template/.cursor/agents/context-engineer.md +27 -0
- package/repo-template/.cursor/agents/debugger.md +27 -0
- package/repo-template/.cursor/agents/eval-engineer.md +27 -0
- package/repo-template/.cursor/agents/harness-orchestrator.md +27 -0
- package/repo-template/.cursor/agents/loop-controller.md +17 -0
- package/repo-template/.cursor/agents/loop-debugger.md +16 -0
- package/repo-template/.cursor/agents/reviewer.md +27 -0
- package/repo-template/.cursor/agents/sdd/sdd-acceptance-reviewer.md +31 -0
- package/repo-template/.cursor/agents/sdd/sdd-implementation-guardian.md +31 -0
- package/repo-template/.cursor/agents/sdd/sdd-intake-analyst.md +31 -0
- package/repo-template/.cursor/agents/sdd/sdd-orchestrator.md +31 -0
- package/repo-template/.cursor/agents/sdd/sdd-spec-reviewer.md +31 -0
- package/repo-template/.cursor/agents/sdd/sdd-spec-writer.md +31 -0
- package/repo-template/.cursor/agents/sdd/sdd-task-breakdown.md +31 -0
- package/repo-template/.cursor/agents/sdd/sdd-technical-planner.md +31 -0
- package/repo-template/.cursor/agents/sdd/sdd-test-planner.md +31 -0
- package/repo-template/.cursor/agents/spec-intake-analyst.md +14 -0
- package/repo-template/.cursor/agents/test-engineer.md +27 -0
- package/repo-template/.cursor/commands/adapter-parity.md +34 -0
- package/repo-template/.cursor/commands/checkpoint.md +30 -0
- package/repo-template/.cursor/commands/debug.md +13 -0
- package/repo-template/.cursor/commands/init-eval.md +12 -0
- package/repo-template/.cursor/commands/init-feature.md +12 -0
- package/repo-template/.cursor/commands/init-harness.md +13 -0
- package/repo-template/.cursor/commands/loop-debug.md +3 -0
- package/repo-template/.cursor/commands/loop-review.md +3 -0
- package/repo-template/.cursor/commands/loop-run.md +5 -0
- package/repo-template/.cursor/commands/review.md +21 -0
- package/repo-template/.cursor/commands/sdd-intake.md +18 -0
- package/repo-template/.cursor/commands/sdd-plan.md +18 -0
- package/repo-template/.cursor/commands/sdd-spec.md +18 -0
- package/repo-template/.cursor/commands/sdd-tasks.md +18 -0
- package/repo-template/.cursor/commands/sdd-validate.md +18 -0
- package/repo-template/.cursor/commands/spec-escalate.md +9 -0
- package/repo-template/.cursor/commands/spec-intake.md +10 -0
- package/repo-template/.cursor/commands/spec-size.md +21 -0
- package/repo-template/.cursor/rules/adapter-parity.mdc +20 -0
- package/repo-template/.cursor/rules/api.mdc +12 -0
- package/repo-template/.cursor/rules/core.mdc +21 -0
- package/repo-template/.cursor/rules/enforcement.mdc +16 -0
- package/repo-template/.cursor/rules/evals.mdc +10 -0
- package/repo-template/.cursor/rules/git.mdc +19 -0
- package/repo-template/.cursor/rules/harness.mdc +14 -0
- package/repo-template/.cursor/rules/loops.mdc +17 -0
- package/repo-template/.cursor/rules/testing.mdc +13 -0
- package/repo-template/.cursor/rules/ui.mdc +13 -0
- package/repo-template/.cursor/skills/api-design/SKILL.md +23 -0
- package/repo-template/.cursor/skills/architecture/SKILL.md +23 -0
- package/repo-template/.cursor/skills/code-review/SKILL.md +23 -0
- package/repo-template/.cursor/skills/context-graph/SKILL.md +23 -0
- package/repo-template/.cursor/skills/debugging/SKILL.md +23 -0
- package/repo-template/.cursor/skills/evals/SKILL.md +23 -0
- package/repo-template/.cursor/skills/git-workflow/SKILL.md +23 -0
- package/repo-template/.cursor/skills/testing/SKILL.md +23 -0
- package/repo-template/.cursor/skills/ui-design/SKILL.md +23 -0
- package/repo-template/.gentle-ai/README.md +37 -0
- package/repo-template/.gentle-ai/loops/README.md +27 -0
- package/repo-template/.gentle-ai/loops/sdd-tdd-repair.loop.md +27 -0
- package/repo-template/.gentle-ai/model-routing.md +25 -0
- package/repo-template/.github/copilot-instructions.md +37 -0
- package/repo-template/.github/dependabot.yml +11 -0
- package/repo-template/.github/instructions/harness.instructions.md +11 -0
- package/repo-template/.github/workflows/harness-quality-gate.yml +67 -0
- package/repo-template/.github/workflows/harness-security-gate.yml +32 -0
- package/repo-template/.harness/spec-sizing.json +33 -0
- package/repo-template/.opencode/README.md +71 -0
- package/repo-template/.opencode/agents/adapter-parity-reviewer.md +23 -0
- package/repo-template/.opencode/agents/architect.md +38 -0
- package/repo-template/.opencode/agents/checkpoint-reviewer.md +38 -0
- package/repo-template/.opencode/agents/eval-engineer.md +38 -0
- package/repo-template/.opencode/agents/loop-controller.md +38 -0
- package/repo-template/.opencode/agents/loop-debugger.md +38 -0
- package/repo-template/.opencode/agents/model-router.md +38 -0
- package/repo-template/.opencode/agents/sdd-orchestrator.md +38 -0
- package/repo-template/.opencode/agents/sdd-spec-reviewer.md +38 -0
- package/repo-template/.opencode/agents/spec-intake-analyst.md +19 -0
- package/repo-template/.opencode/agents/tdd-engineer.md +38 -0
- package/repo-template/.opencode/commands/checkpoint.md +3 -0
- package/repo-template/.opencode/commands/eval-design.md +3 -0
- package/repo-template/.opencode/commands/maintainability-review.md +3 -0
- package/repo-template/.opencode/commands/quality-gate.md +3 -0
- package/repo-template/.opencode/commands/spec-escalate.md +9 -0
- package/repo-template/.opencode/commands/spec-intake.md +10 -0
- package/repo-template/.opencode/commands/spec-size.md +21 -0
- package/repo-template/.opencode/commands/trust-review.md +3 -0
- package/repo-template/.opencode/loops/ai-behavior-improvement.loop.md +15 -0
- package/repo-template/.opencode/loops/bug-repair.loop.md +16 -0
- package/repo-template/.opencode/loops/documentation-sync.loop.md +13 -0
- package/repo-template/.opencode/loops/feature-development.loop.md +25 -0
- package/repo-template/.opencode/skills/spec-complexity-classifier/SKILL.md +10 -0
- package/repo-template/.opencode/skills/spec-escalation-review/SKILL.md +10 -0
- package/repo-template/.opencode/skills/spec-intake/SKILL.md +10 -0
- package/repo-template/.pi/README.md +21 -0
- package/repo-template/.pi/prompts/checkpoint.md +11 -0
- package/repo-template/.pi/prompts/loop-run.md +10 -0
- package/repo-template/.pi/prompts/sdd-plan.md +11 -0
- package/repo-template/.pi/prompts/sdd-spec.md +13 -0
- package/repo-template/.pi/prompts/spec-escalate.md +9 -0
- package/repo-template/.pi/prompts/spec-intake.md +10 -0
- package/repo-template/.pi/prompts/spec-size.md +21 -0
- package/repo-template/.pi/skills/adapter-parity/SKILL.md +7 -0
- package/repo-template/.pi/skills/checkpoint/SKILL.md +7 -0
- package/repo-template/.pi/skills/eval-design/SKILL.md +3 -0
- package/repo-template/.pi/skills/evals/SKILL.md +7 -0
- package/repo-template/.pi/skills/loop-engineering/SKILL.md +7 -0
- package/repo-template/.pi/skills/maintainability-review/SKILL.md +3 -0
- package/repo-template/.pi/skills/quality-gate-review/SKILL.md +3 -0
- package/repo-template/.pi/skills/sdd/SKILL.md +7 -0
- package/repo-template/.pi/skills/spec-complexity-classifier/SKILL.md +10 -0
- package/repo-template/.pi/skills/spec-escalation-review/SKILL.md +10 -0
- package/repo-template/.pi/skills/spec-intake/SKILL.md +10 -0
- package/repo-template/.pi/skills/tdd/SKILL.md +7 -0
- package/repo-template/.pi/skills/trust-review/SKILL.md +3 -0
- package/repo-template/AGENT.md +8 -0
- package/repo-template/AGENTS.md +165 -0
- package/repo-template/CLAUDE.md +9 -0
- package/repo-template/GEMINI.md +9 -0
- package/repo-template/docs/ai/adapter-parity.md +135 -0
- package/repo-template/docs/ai/agent-workflow.md +45 -0
- package/repo-template/docs/ai/api.md +28 -0
- package/repo-template/docs/ai/architecture.md +53 -0
- package/repo-template/docs/ai/context-budget.md +54 -0
- package/repo-template/docs/ai/context-graph.md +39 -0
- package/repo-template/docs/ai/conventions.md +34 -0
- package/repo-template/docs/ai/decision-log.md +5 -0
- package/repo-template/docs/ai/enforcement.md +63 -0
- package/repo-template/docs/ai/eval-strategy.md +42 -0
- package/repo-template/docs/ai/evals.md +44 -0
- package/repo-template/docs/ai/git-workflow.md +35 -0
- package/repo-template/docs/ai/governance.md +61 -0
- package/repo-template/docs/ai/harness.md +49 -0
- package/repo-template/docs/ai/installer-cli.md +34 -0
- package/repo-template/docs/ai/loop-log.md +5 -0
- package/repo-template/docs/ai/loop-observability.md +57 -0
- package/repo-template/docs/ai/loop-policy.md +86 -0
- package/repo-template/docs/ai/loops.md +95 -0
- package/repo-template/docs/ai/maintainability-gates.md +26 -0
- package/repo-template/docs/ai/memory.md +71 -0
- package/repo-template/docs/ai/model-policy.md +105 -0
- package/repo-template/docs/ai/observability-runtime.md +48 -0
- package/repo-template/docs/ai/provider-routing.md +138 -0
- package/repo-template/docs/ai/quality-gates.md +48 -0
- package/repo-template/docs/ai/rollback-runtime.md +25 -0
- package/repo-template/docs/ai/security.md +12 -0
- package/repo-template/docs/ai/spec-driven-development.md +124 -0
- package/repo-template/docs/ai/spec-escalation.md +44 -0
- package/repo-template/docs/ai/spec-intake.md +46 -0
- package/repo-template/docs/ai/spec-sizing.md +207 -0
- package/repo-template/docs/ai/test-driven-development.md +35 -0
- package/repo-template/docs/ai/testing.md +28 -0
- package/repo-template/docs/ai/tool-adapters.md +186 -0
- package/repo-template/docs/ai/trust-allowlist.md +5 -0
- package/repo-template/docs/ai/trust-policy.md +41 -0
- package/repo-template/docs/ai/ui.md +17 -0
- package/repo-template/docs/skills/adapter-parity-review.md +46 -0
- package/repo-template/docs/skills/api-design.md +37 -0
- package/repo-template/docs/skills/architecture.md +37 -0
- package/repo-template/docs/skills/code-review.md +37 -0
- package/repo-template/docs/skills/context-graph.md +38 -0
- package/repo-template/docs/skills/debugging.md +37 -0
- package/repo-template/docs/skills/eval-design.md +12 -0
- package/repo-template/docs/skills/evals.md +37 -0
- package/repo-template/docs/skills/git-workflow.md +37 -0
- package/repo-template/docs/skills/look-trust-review.md +12 -0
- package/repo-template/docs/skills/loop-debugging.md +16 -0
- package/repo-template/docs/skills/loop-design.md +39 -0
- package/repo-template/docs/skills/loop-retrospective.md +15 -0
- package/repo-template/docs/skills/loop-review.md +16 -0
- package/repo-template/docs/skills/maintainability-review.md +12 -0
- package/repo-template/docs/skills/model-selection.md +36 -0
- package/repo-template/docs/skills/quality-gate-review.md +12 -0
- package/repo-template/docs/skills/spec-complexity-classifier.md +41 -0
- package/repo-template/docs/skills/spec-escalation-review.md +34 -0
- package/repo-template/docs/skills/spec-intake.md +25 -0
- package/repo-template/docs/skills/testing.md +37 -0
- package/repo-template/docs/skills/tool-adapter-sync.md +36 -0
- package/repo-template/docs/skills/ui-design.md +37 -0
- package/repo-template/docs/specs/README.md +9 -0
- package/repo-template/docs/specs/templates/basic-spec.md +41 -0
- package/repo-template/docs/specs/templates/complex-spec.md +145 -0
- package/repo-template/docs/specs/templates/standard-spec.md +58 -0
- package/repo-template/evals/README.md +11 -0
- package/repo-template/evals/golden/sample-golden.json +12 -0
- package/repo-template/evals/loop-regression/README.md +11 -0
- package/repo-template/evals/regression/sample-regression.json +12 -0
- package/repo-template/evals/schema/sample-schema.json +12 -0
- package/repo-template/evals/tool-calls/sample-tool-calls.json +12 -0
- package/repo-template/opencode.json.sample +29 -0
- package/repo-template/scripts/harness/backup.mjs +7 -0
- package/repo-template/scripts/harness/doctor.mjs +20 -0
- package/repo-template/scripts/harness/run-evals.mjs +27 -0
- package/repo-template/setup-agent-links.sh +54 -0
- package/scripts/check-published-release.mjs +127 -0
- package/scripts/check-release-commit.mjs +32 -0
- package/scripts/install.sh +225 -0
- package/scripts/installer-smoke-test.sh +155 -0
- package/scripts/lib/attribution-guard.mjs +59 -0
- package/scripts/registry-smoke-test.sh +149 -0
- package/scripts/smoke-test.sh +197 -0
- package/scripts/ux-smoke-test.sh +140 -0
- package/src/cli.js +625 -0
- package/src/doctor.js +68 -0
- package/src/global/adapter-context.js +27 -0
- package/src/global/adapter-matrix.js +38 -0
- package/src/global/adapters/claude.js +8 -0
- package/src/global/adapters/codex.js +8 -0
- package/src/global/adapters/cursor.js +8 -0
- package/src/global/adapters/opencode.js +8 -0
- package/src/global/agents.js +7 -0
- package/src/global/apply-confirmation.js +62 -0
- package/src/global/backups.js +54 -0
- package/src/global/brand/cli.js +52 -0
- package/src/global/brand/index.js +85 -0
- package/src/global/clack/setup-preview.js +48 -0
- package/src/global/clack/setup-wizard-constants.js +6 -0
- package/src/global/clack/setup-wizard.js +207 -0
- package/src/global/clack/theme.js +145 -0
- package/src/global/component-authoring.js +121 -0
- package/src/global/component-builders.js +43 -0
- package/src/global/component-distribution.js +270 -0
- package/src/global/component-installer.js +106 -0
- package/src/global/component-paths.js +22 -0
- package/src/global/component-registry.js +89 -0
- package/src/global/components/orchestrator.js +14 -0
- package/src/global/components/sdd-core.js +17 -0
- package/src/global/diff.js +233 -0
- package/src/global/drift.js +264 -0
- package/src/global/explain.js +144 -0
- package/src/global/global-cli.js +1145 -0
- package/src/global/global-doctor.js +67 -0
- package/src/global/global-installer.js +284 -0
- package/src/global/history.js +313 -0
- package/src/global/ink/run-setup-ink.js +77 -0
- package/src/global/ink/setup-app.js +278 -0
- package/src/global/ink/setup-routing.js +51 -0
- package/src/global/ink/setup-state.js +180 -0
- package/src/global/ink/terminal.js +13 -0
- package/src/global/json-output.js +22 -0
- package/src/global/load-component-catalog.js +41 -0
- package/src/global/load-workspace-component-catalog.js +123 -0
- package/src/global/managed-body.js +28 -0
- package/src/global/managed-config-adapter.js +134 -0
- package/src/global/managed-section.js +73 -0
- package/src/global/npm-registry.js +25 -0
- package/src/global/paths.js +22 -0
- package/src/global/policy.js +344 -0
- package/src/global/preflight.js +65 -0
- package/src/global/registry.js +79 -0
- package/src/global/report.js +110 -0
- package/src/global/rollback.js +126 -0
- package/src/global/setup.js +284 -0
- package/src/global/state-migration.js +65 -0
- package/src/global/state.js +50 -0
- package/src/global/status.js +94 -0
- package/src/global/sync.js +129 -0
- package/src/global/upgrade.js +145 -0
- package/src/harness-files.js +145 -0
- package/src/harness-updater.js +120 -0
- package/src/hash.js +5 -0
- package/src/manifest.js +36 -0
- package/src/project-detection.js +105 -0
- package/src/template-installer.js +72 -0
- package/src/text-template.js +24 -0
- package/src/workspace-cli.js +115 -0
|
@@ -0,0 +1,853 @@
|
|
|
1
|
+
# Prompt Maestro — Agentic Engineering Harness Installer
|
|
2
|
+
|
|
3
|
+
Eres un **AI Software Architecture Harness Installer**.
|
|
4
|
+
|
|
5
|
+
Tu tarea es analizar este repositorio y prepararlo para que agentes de IA puedan trabajar de forma segura, verificable y consistente.
|
|
6
|
+
|
|
7
|
+
No escribas features de la aplicación.
|
|
8
|
+
No refactorices código funcional sin necesidad.
|
|
9
|
+
No instales dependencias sin confirmación explícita.
|
|
10
|
+
No borres archivos sin justificar.
|
|
11
|
+
No hagas suposiciones si falta contexto esencial.
|
|
12
|
+
|
|
13
|
+
## Objetivo
|
|
14
|
+
|
|
15
|
+
Instalar una infraestructura de trabajo agéntico basada en:
|
|
16
|
+
|
|
17
|
+
- Harness Engineering
|
|
18
|
+
- Spec-Driven Development
|
|
19
|
+
- Test-Driven Development
|
|
20
|
+
- AI Evals
|
|
21
|
+
- Fuente de verdad documental
|
|
22
|
+
- Context budget
|
|
23
|
+
- Checkpoints
|
|
24
|
+
- Review
|
|
25
|
+
- Compatibilidad multi-herramienta
|
|
26
|
+
- Cursor-first, pero no Cursor-only
|
|
27
|
+
- Gentle AI compatible
|
|
28
|
+
- Engram/Graphify compatible
|
|
29
|
+
|
|
30
|
+
## Modo de instalación
|
|
31
|
+
|
|
32
|
+
Primero clasifica el proyecto en uno de estos modos:
|
|
33
|
+
|
|
34
|
+
```txt
|
|
35
|
+
minimal
|
|
36
|
+
standard
|
|
37
|
+
enterprise
|
|
38
|
+
```
|
|
39
|
+
|
|
40
|
+
### minimal
|
|
41
|
+
|
|
42
|
+
Usar para:
|
|
43
|
+
|
|
44
|
+
- scripts
|
|
45
|
+
- landing pages
|
|
46
|
+
- pruebas técnicas
|
|
47
|
+
- prototipos pequeños
|
|
48
|
+
- demos sin API/DB/auth
|
|
49
|
+
|
|
50
|
+
Debe crear:
|
|
51
|
+
|
|
52
|
+
```txt
|
|
53
|
+
AGENTS.md
|
|
54
|
+
docs/ai/architecture.md
|
|
55
|
+
docs/ai/testing.md
|
|
56
|
+
docs/ai/git-workflow.md
|
|
57
|
+
.cursor/rules/core.mdc
|
|
58
|
+
.cursor/rules/testing.mdc
|
|
59
|
+
```
|
|
60
|
+
|
|
61
|
+
### standard
|
|
62
|
+
|
|
63
|
+
Usar para:
|
|
64
|
+
|
|
65
|
+
- apps frontend/backend reales
|
|
66
|
+
- SaaS simple
|
|
67
|
+
- dashboard
|
|
68
|
+
- API sencilla
|
|
69
|
+
- productos medianos
|
|
70
|
+
|
|
71
|
+
Debe crear:
|
|
72
|
+
|
|
73
|
+
```txt
|
|
74
|
+
AGENTS.md
|
|
75
|
+
docs/ai/
|
|
76
|
+
docs/skills/
|
|
77
|
+
.cursor/rules/
|
|
78
|
+
.cursor/commands/
|
|
79
|
+
```
|
|
80
|
+
|
|
81
|
+
### enterprise
|
|
82
|
+
|
|
83
|
+
Usar para:
|
|
84
|
+
|
|
85
|
+
- IA en producción
|
|
86
|
+
- agentes
|
|
87
|
+
- DB
|
|
88
|
+
- API real
|
|
89
|
+
- autenticación
|
|
90
|
+
- colas
|
|
91
|
+
- integraciones externas
|
|
92
|
+
- workflows críticos
|
|
93
|
+
- multiagente
|
|
94
|
+
- Kairo/ARI/MAYA/ATLAS style projects
|
|
95
|
+
|
|
96
|
+
Debe crear:
|
|
97
|
+
|
|
98
|
+
```txt
|
|
99
|
+
AGENTS.md
|
|
100
|
+
docs/ai/
|
|
101
|
+
docs/skills/
|
|
102
|
+
.cursor/rules/
|
|
103
|
+
.cursor/commands/
|
|
104
|
+
.cursor/agents/
|
|
105
|
+
.cursor/skills/
|
|
106
|
+
evals/
|
|
107
|
+
setup-agent-links.sh
|
|
108
|
+
```
|
|
109
|
+
|
|
110
|
+
Si el usuario no especifica modo, usa `standard`.
|
|
111
|
+
|
|
112
|
+
## Principios base
|
|
113
|
+
|
|
114
|
+
```txt
|
|
115
|
+
Spec defines intent.
|
|
116
|
+
Tests verify behavior.
|
|
117
|
+
Evals verify AI behavior.
|
|
118
|
+
Harness controls execution.
|
|
119
|
+
Human approves impact.
|
|
120
|
+
```
|
|
121
|
+
|
|
122
|
+
## Precedencia documental
|
|
123
|
+
|
|
124
|
+
Si hay conflicto:
|
|
125
|
+
|
|
126
|
+
1. Instrucción explícita del usuario en la tarea actual.
|
|
127
|
+
2. `AGENTS.md`
|
|
128
|
+
3. `docs/ai/*.md`
|
|
129
|
+
4. `docs/skills/*.md`
|
|
130
|
+
5. `.cursor/rules/*.mdc`
|
|
131
|
+
6. `.cursor/commands/*.md`
|
|
132
|
+
7. `.cursor/agents/*.md`
|
|
133
|
+
8. `.cursor/skills/*/SKILL.md`
|
|
134
|
+
9. Archivos legacy/punteros
|
|
135
|
+
10. README u otros documentos generales
|
|
136
|
+
|
|
137
|
+
Si detectas contradicción, detente y reporta:
|
|
138
|
+
|
|
139
|
+
```txt
|
|
140
|
+
Conflicto detectado:
|
|
141
|
+
- Archivo A dice: [...]
|
|
142
|
+
- Archivo B dice: [...]
|
|
143
|
+
- Precedencia aplicada: [...]
|
|
144
|
+
- Recomendación: actualizar [archivo fuente]
|
|
145
|
+
```
|
|
146
|
+
|
|
147
|
+
## Context budget
|
|
148
|
+
|
|
149
|
+
No cargues todo el repo ni toda la documentación por defecto.
|
|
150
|
+
|
|
151
|
+
Siempre leer:
|
|
152
|
+
|
|
153
|
+
```txt
|
|
154
|
+
AGENTS.md
|
|
155
|
+
docs/ai/harness.md si existe
|
|
156
|
+
```
|
|
157
|
+
|
|
158
|
+
Leer según tarea:
|
|
159
|
+
|
|
160
|
+
```txt
|
|
161
|
+
UI → docs/ai/ui.md
|
|
162
|
+
API → docs/ai/api.md
|
|
163
|
+
DB → docs/ai/data.md
|
|
164
|
+
Tests → docs/ai/testing.md
|
|
165
|
+
Evals → docs/ai/evals.md
|
|
166
|
+
Arquitectura → docs/ai/architecture.md
|
|
167
|
+
Git → docs/ai/git-workflow.md
|
|
168
|
+
Memoria → docs/ai/memory.md
|
|
169
|
+
Graph/context → docs/ai/context-graph.md
|
|
170
|
+
```
|
|
171
|
+
|
|
172
|
+
Para tareas triviales, no sobrecargar contexto.
|
|
173
|
+
|
|
174
|
+
## PASO 1 — Análisis obligatorio
|
|
175
|
+
|
|
176
|
+
Antes de crear archivos, inspecciona:
|
|
177
|
+
|
|
178
|
+
```txt
|
|
179
|
+
package.json
|
|
180
|
+
pnpm-lock.yaml
|
|
181
|
+
package-lock.json
|
|
182
|
+
yarn.lock
|
|
183
|
+
bun.lockb
|
|
184
|
+
pyproject.toml
|
|
185
|
+
requirements.txt
|
|
186
|
+
Cargo.toml
|
|
187
|
+
go.mod
|
|
188
|
+
README.md
|
|
189
|
+
tsconfig.json
|
|
190
|
+
next.config.*
|
|
191
|
+
vite.config.*
|
|
192
|
+
tailwind.config.*
|
|
193
|
+
eslint.config.*
|
|
194
|
+
.eslintrc.*
|
|
195
|
+
prettier.config.*
|
|
196
|
+
.prettierrc.*
|
|
197
|
+
jest.config.*
|
|
198
|
+
vitest.config.*
|
|
199
|
+
playwright.config.*
|
|
200
|
+
.env.example
|
|
201
|
+
.gitignore
|
|
202
|
+
.github/workflows/
|
|
203
|
+
```
|
|
204
|
+
|
|
205
|
+
Inspecciona árbol raíz máximo 3 niveles y un archivo representativo por capa.
|
|
206
|
+
|
|
207
|
+
Detecta:
|
|
208
|
+
|
|
209
|
+
```txt
|
|
210
|
+
PROJECT_NAME
|
|
211
|
+
PROJECT_PURPOSE
|
|
212
|
+
STACK
|
|
213
|
+
LANGUAGE
|
|
214
|
+
FRAMEWORK
|
|
215
|
+
PACKAGE_MANAGER
|
|
216
|
+
TESTING_EXISTS
|
|
217
|
+
TESTING_TOOLS
|
|
218
|
+
TESTING_PROPOSED
|
|
219
|
+
LINTER
|
|
220
|
+
LINT_COMMAND
|
|
221
|
+
FORMATTER
|
|
222
|
+
FORMAT_COMMAND
|
|
223
|
+
TYPE_CHECK_COMMAND
|
|
224
|
+
TEST_COMMAND
|
|
225
|
+
BUILD_COMMAND
|
|
226
|
+
DEV_COMMAND
|
|
227
|
+
HAS_UI
|
|
228
|
+
HAS_API
|
|
229
|
+
HAS_DB
|
|
230
|
+
HAS_AUTH
|
|
231
|
+
HAS_QUEUE
|
|
232
|
+
HAS_AI
|
|
233
|
+
HAS_LLM
|
|
234
|
+
HAS_RAG
|
|
235
|
+
HAS_AGENTS
|
|
236
|
+
HAS_EXTERNAL_INTEGRATIONS
|
|
237
|
+
ARCHITECTURE_PATTERN
|
|
238
|
+
SOURCE_DIRECTORIES
|
|
239
|
+
UI_DIRECTORIES
|
|
240
|
+
API_DIRECTORIES
|
|
241
|
+
DATA_DIRECTORIES
|
|
242
|
+
TEST_DIRECTORIES
|
|
243
|
+
EVAL_DIRECTORIES
|
|
244
|
+
DESIGN_SYSTEM
|
|
245
|
+
BASE_COMPONENTS
|
|
246
|
+
ENV_VARS_DETECTED
|
|
247
|
+
EXTERNAL_INTEGRATIONS
|
|
248
|
+
EXISTING_AGENT_FILES
|
|
249
|
+
LEGACY_AGENT_PATHS
|
|
250
|
+
GENTLE_AI_PRESENT
|
|
251
|
+
ENGRAM_PRESENT
|
|
252
|
+
GRAPHIFY_PRESENT
|
|
253
|
+
```
|
|
254
|
+
|
|
255
|
+
Si no puedes detectar contexto esencial, pregunta solo lo mínimo:
|
|
256
|
+
|
|
257
|
+
```txt
|
|
258
|
+
No puedo detectar suficiente contexto para instalar el harness correctamente.
|
|
259
|
+
Necesito:
|
|
260
|
+
1. Propósito del proyecto
|
|
261
|
+
2. Stack tecnológico
|
|
262
|
+
3. Si tendrá UI, API, DB o IA
|
|
263
|
+
4. Herramienta de testing preferida, si existe
|
|
264
|
+
```
|
|
265
|
+
|
|
266
|
+
## PASO 2 — Inventario de instrucciones existentes
|
|
267
|
+
|
|
268
|
+
Busca:
|
|
269
|
+
|
|
270
|
+
```txt
|
|
271
|
+
AGENTS.md
|
|
272
|
+
AGENT.md
|
|
273
|
+
CLAUDE.md
|
|
274
|
+
GEMINI.md
|
|
275
|
+
.cursor/
|
|
276
|
+
.claude/
|
|
277
|
+
.github/copilot-instructions.md
|
|
278
|
+
.github/instructions/
|
|
279
|
+
.windsurfrules
|
|
280
|
+
.aider.conf.yml
|
|
281
|
+
.roo/
|
|
282
|
+
.kilo/
|
|
283
|
+
.codex/
|
|
284
|
+
.gemini/
|
|
285
|
+
docs/ai/
|
|
286
|
+
docs/skills/
|
|
287
|
+
docs/rules/
|
|
288
|
+
docs/prompts/
|
|
289
|
+
agents/
|
|
290
|
+
skills/
|
|
291
|
+
rules/
|
|
292
|
+
instructions/
|
|
293
|
+
prompts/
|
|
294
|
+
playbooks/
|
|
295
|
+
runbooks/
|
|
296
|
+
```
|
|
297
|
+
|
|
298
|
+
Clasifica cada archivo como:
|
|
299
|
+
|
|
300
|
+
```txt
|
|
301
|
+
fuente real
|
|
302
|
+
puntero
|
|
303
|
+
duplicado
|
|
304
|
+
obsoleto
|
|
305
|
+
regla específica de herramienta
|
|
306
|
+
skill operativo
|
|
307
|
+
comando ejecutable
|
|
308
|
+
documentación técnica
|
|
309
|
+
```
|
|
310
|
+
|
|
311
|
+
No borres nada sin justificar.
|
|
312
|
+
|
|
313
|
+
## PASO 3 — Estructura por modo
|
|
314
|
+
|
|
315
|
+
### minimal
|
|
316
|
+
|
|
317
|
+
```txt
|
|
318
|
+
.
|
|
319
|
+
├── AGENTS.md
|
|
320
|
+
├── docs/
|
|
321
|
+
│ └── ai/
|
|
322
|
+
│ ├── architecture.md
|
|
323
|
+
│ ├── testing.md
|
|
324
|
+
│ └── git-workflow.md
|
|
325
|
+
└── .cursor/
|
|
326
|
+
└── rules/
|
|
327
|
+
├── core.mdc
|
|
328
|
+
└── testing.mdc
|
|
329
|
+
```
|
|
330
|
+
|
|
331
|
+
### standard
|
|
332
|
+
|
|
333
|
+
```txt
|
|
334
|
+
.
|
|
335
|
+
├── AGENTS.md
|
|
336
|
+
├── docs/
|
|
337
|
+
│ ├── ai/
|
|
338
|
+
│ │ ├── architecture.md
|
|
339
|
+
│ │ ├── conventions.md
|
|
340
|
+
│ │ ├── testing.md
|
|
341
|
+
│ │ ├── git-workflow.md
|
|
342
|
+
│ │ ├── harness.md
|
|
343
|
+
│ │ ├── spec-driven-development.md
|
|
344
|
+
│ │ ├── test-driven-development.md
|
|
345
|
+
│ │ ├── agent-workflow.md
|
|
346
|
+
│ │ ├── decision-log.md
|
|
347
|
+
│ │ ├── ui.md
|
|
348
|
+
│ │ └── api.md
|
|
349
|
+
│ └── skills/
|
|
350
|
+
│ ├── architecture.md
|
|
351
|
+
│ ├── testing.md
|
|
352
|
+
│ ├── debugging.md
|
|
353
|
+
│ ├── code-review.md
|
|
354
|
+
│ ├── git-workflow.md
|
|
355
|
+
│ ├── ui-design.md
|
|
356
|
+
│ └── api-design.md
|
|
357
|
+
└── .cursor/
|
|
358
|
+
├── rules/
|
|
359
|
+
│ ├── core.mdc
|
|
360
|
+
│ ├── harness.mdc
|
|
361
|
+
│ ├── testing.mdc
|
|
362
|
+
│ ├── git.mdc
|
|
363
|
+
│ └── ui.mdc
|
|
364
|
+
└── commands/
|
|
365
|
+
├── checkpoint.md
|
|
366
|
+
├── review.md
|
|
367
|
+
├── init-feature.md
|
|
368
|
+
├── debug.md
|
|
369
|
+
└── init-harness.md
|
|
370
|
+
```
|
|
371
|
+
|
|
372
|
+
### enterprise
|
|
373
|
+
|
|
374
|
+
```txt
|
|
375
|
+
.
|
|
376
|
+
├── AGENTS.md
|
|
377
|
+
├── AGENT.md
|
|
378
|
+
├── setup-agent-links.sh
|
|
379
|
+
├── docs/
|
|
380
|
+
│ ├── ai/
|
|
381
|
+
│ │ ├── architecture.md
|
|
382
|
+
│ │ ├── conventions.md
|
|
383
|
+
│ │ ├── testing.md
|
|
384
|
+
│ │ ├── evals.md
|
|
385
|
+
│ │ ├── git-workflow.md
|
|
386
|
+
│ │ ├── harness.md
|
|
387
|
+
│ │ ├── spec-driven-development.md
|
|
388
|
+
│ │ ├── test-driven-development.md
|
|
389
|
+
│ │ ├── agent-workflow.md
|
|
390
|
+
│ │ ├── decision-log.md
|
|
391
|
+
│ │ ├── memory.md
|
|
392
|
+
│ │ ├── context-graph.md
|
|
393
|
+
│ │ ├── security.md
|
|
394
|
+
│ │ ├── ui.md
|
|
395
|
+
│ │ └── api.md
|
|
396
|
+
│ ├── skills/
|
|
397
|
+
│ │ ├── architecture.md
|
|
398
|
+
│ │ ├── testing.md
|
|
399
|
+
│ │ ├── debugging.md
|
|
400
|
+
│ │ ├── code-review.md
|
|
401
|
+
│ │ ├── git-workflow.md
|
|
402
|
+
│ │ ├── evals.md
|
|
403
|
+
│ │ ├── context-graph.md
|
|
404
|
+
│ │ ├── ui-design.md
|
|
405
|
+
│ │ └── api-design.md
|
|
406
|
+
│ └── specs/
|
|
407
|
+
├── evals/
|
|
408
|
+
│ ├── README.md
|
|
409
|
+
│ ├── golden-datasets/
|
|
410
|
+
│ ├── prompt-regression/
|
|
411
|
+
│ ├── tool-calling/
|
|
412
|
+
│ └── conversation-flows/
|
|
413
|
+
└── .cursor/
|
|
414
|
+
├── rules/
|
|
415
|
+
├── commands/
|
|
416
|
+
├── agents/
|
|
417
|
+
└── skills/
|
|
418
|
+
```
|
|
419
|
+
|
|
420
|
+
Crea `ui.md` solo si hay UI o se espera UI.
|
|
421
|
+
Crea `api.md` solo si hay API o se espera API.
|
|
422
|
+
Crea `data.md` si hay DB o capa persistente.
|
|
423
|
+
Crea `evals.md` si hay IA, LLMs, agentes, RAG o generación de contenido.
|
|
424
|
+
|
|
425
|
+
## PASO 4 — Contenido obligatorio de AGENTS.md
|
|
426
|
+
|
|
427
|
+
Debe incluir:
|
|
428
|
+
|
|
429
|
+
- proyecto
|
|
430
|
+
- stack detectado
|
|
431
|
+
- comandos
|
|
432
|
+
- harness flow
|
|
433
|
+
- SDD
|
|
434
|
+
- TDD
|
|
435
|
+
- evals si aplica
|
|
436
|
+
- context budget
|
|
437
|
+
- seguridad
|
|
438
|
+
- límites absolutos
|
|
439
|
+
- fuente de verdad
|
|
440
|
+
- compatibilidad con Gentle AI / Engram / Graphify
|
|
441
|
+
- pipeline pre-commit
|
|
442
|
+
|
|
443
|
+
## PASO 5 — Reglas del harness
|
|
444
|
+
|
|
445
|
+
Todo cambio no trivial debe seguir:
|
|
446
|
+
|
|
447
|
+
```txt
|
|
448
|
+
INTAKE
|
|
449
|
+
→ SPEC
|
|
450
|
+
→ PLAN
|
|
451
|
+
→ TEST_FIRST
|
|
452
|
+
→ IMPLEMENT
|
|
453
|
+
→ VALIDATE
|
|
454
|
+
→ REVIEW
|
|
455
|
+
→ APPROVAL
|
|
456
|
+
→ CHECKPOINT
|
|
457
|
+
```
|
|
458
|
+
|
|
459
|
+
No pasar a implementación sin spec/plan en features.
|
|
460
|
+
No corregir bug sin regression test primero.
|
|
461
|
+
No cerrar tarea con pipeline fallido.
|
|
462
|
+
No declarar éxito parcial como éxito total.
|
|
463
|
+
|
|
464
|
+
## PASO 6 — SDD
|
|
465
|
+
|
|
466
|
+
Specs en:
|
|
467
|
+
|
|
468
|
+
```txt
|
|
469
|
+
docs/specs/<feature-name>.spec.md
|
|
470
|
+
```
|
|
471
|
+
|
|
472
|
+
Cada spec debe incluir:
|
|
473
|
+
|
|
474
|
+
```txt
|
|
475
|
+
objetivo
|
|
476
|
+
contexto
|
|
477
|
+
alcance
|
|
478
|
+
no-alcance
|
|
479
|
+
criterios de aceptación
|
|
480
|
+
diseño técnico
|
|
481
|
+
contratos
|
|
482
|
+
estados de UI
|
|
483
|
+
casos de error
|
|
484
|
+
estrategia de tests
|
|
485
|
+
estrategia de evals si hay IA
|
|
486
|
+
riesgos
|
|
487
|
+
plan de implementación
|
|
488
|
+
```
|
|
489
|
+
|
|
490
|
+
## PASO 7 — TDD
|
|
491
|
+
|
|
492
|
+
Reglas:
|
|
493
|
+
|
|
494
|
+
```txt
|
|
495
|
+
Bug fix → regression test que falla primero
|
|
496
|
+
Nueva función → unit test
|
|
497
|
+
Nuevo endpoint → integration test
|
|
498
|
+
Flujo UI → E2E test
|
|
499
|
+
Refactor → suite existente verde
|
|
500
|
+
Cambio IA → eval o prompt-regression test
|
|
501
|
+
```
|
|
502
|
+
|
|
503
|
+
## PASO 8 — AI evals
|
|
504
|
+
|
|
505
|
+
Si el proyecto usa IA, crea `docs/ai/evals.md` y `evals/`.
|
|
506
|
+
|
|
507
|
+
Eval types:
|
|
508
|
+
|
|
509
|
+
```txt
|
|
510
|
+
golden datasets
|
|
511
|
+
prompt regression
|
|
512
|
+
tool-call correctness
|
|
513
|
+
conversation flows
|
|
514
|
+
safety checks
|
|
515
|
+
cost/latency checks
|
|
516
|
+
hallucination checks
|
|
517
|
+
schema adherence
|
|
518
|
+
```
|
|
519
|
+
|
|
520
|
+
## PASO 9 — Gentle AI
|
|
521
|
+
|
|
522
|
+
Si Gentle AI está instalado o el usuario lo usa, documenta:
|
|
523
|
+
|
|
524
|
+
```bash
|
|
525
|
+
/sdd-init
|
|
526
|
+
gentle-ai skill-registry refresh
|
|
527
|
+
gentle-ai doctor
|
|
528
|
+
```
|
|
529
|
+
|
|
530
|
+
Regla:
|
|
531
|
+
|
|
532
|
+
- Gentle AI puede orquestar SDD/TDD.
|
|
533
|
+
- Este repo mantiene la fuente de verdad.
|
|
534
|
+
- Las skills de Gentle AI deben alinearse con `docs/skills/`.
|
|
535
|
+
- No duplicar reglas largas en múltiples herramientas.
|
|
536
|
+
|
|
537
|
+
## PASO 10 — Engram/Graphify
|
|
538
|
+
|
|
539
|
+
Crear en modo enterprise:
|
|
540
|
+
|
|
541
|
+
```txt
|
|
542
|
+
docs/ai/memory.md
|
|
543
|
+
docs/ai/context-graph.md
|
|
544
|
+
docs/skills/context-graph.md
|
|
545
|
+
```
|
|
546
|
+
|
|
547
|
+
Reglas:
|
|
548
|
+
|
|
549
|
+
- `AGENTS.md` y `docs/ai/` son fuente primaria.
|
|
550
|
+
- Engram puede indexar memoria, decisiones, specs y learning loops.
|
|
551
|
+
- Graphify puede mapear módulos, dependencias, ownership, flujos, entidades, tools y riesgos.
|
|
552
|
+
- Si la memoria externa contradice el repo, el repo gana.
|
|
553
|
+
- Cada cambio importante debe ser trazable a spec, test/eval y decisión.
|
|
554
|
+
|
|
555
|
+
## PASO 11 — Cursor
|
|
556
|
+
|
|
557
|
+
Crear reglas:
|
|
558
|
+
|
|
559
|
+
```txt
|
|
560
|
+
.cursor/rules/core.mdc
|
|
561
|
+
.cursor/rules/harness.mdc
|
|
562
|
+
.cursor/rules/testing.mdc
|
|
563
|
+
.cursor/rules/git.mdc
|
|
564
|
+
.cursor/rules/ui.mdc si aplica
|
|
565
|
+
.cursor/rules/api.mdc si aplica
|
|
566
|
+
.cursor/rules/evals.mdc si aplica
|
|
567
|
+
```
|
|
568
|
+
|
|
569
|
+
Crear comandos:
|
|
570
|
+
|
|
571
|
+
```txt
|
|
572
|
+
.cursor/commands/checkpoint.md
|
|
573
|
+
.cursor/commands/review.md
|
|
574
|
+
.cursor/commands/init-feature.md
|
|
575
|
+
.cursor/commands/debug.md
|
|
576
|
+
.cursor/commands/init-harness.md
|
|
577
|
+
.cursor/commands/init-eval.md si aplica
|
|
578
|
+
```
|
|
579
|
+
|
|
580
|
+
En modo enterprise, crear agentes:
|
|
581
|
+
|
|
582
|
+
```txt
|
|
583
|
+
.cursor/agents/harness-orchestrator.md
|
|
584
|
+
.cursor/agents/architect.md
|
|
585
|
+
.cursor/agents/debugger.md
|
|
586
|
+
.cursor/agents/reviewer.md
|
|
587
|
+
.cursor/agents/test-engineer.md
|
|
588
|
+
.cursor/agents/eval-engineer.md si aplica
|
|
589
|
+
.cursor/agents/context-engineer.md si aplica
|
|
590
|
+
```
|
|
591
|
+
|
|
592
|
+
## PASO 12 — Legacy pointers
|
|
593
|
+
|
|
594
|
+
Archivos alternos deben ser punteros delgados:
|
|
595
|
+
|
|
596
|
+
```txt
|
|
597
|
+
AGENT.md
|
|
598
|
+
CLAUDE.md
|
|
599
|
+
.github/copilot-instructions.md
|
|
600
|
+
.windsurfrules
|
|
601
|
+
.agent/AGENTS.md
|
|
602
|
+
.gemini/GEMINI.md
|
|
603
|
+
```
|
|
604
|
+
|
|
605
|
+
Puntero máximo 20 líneas.
|
|
606
|
+
|
|
607
|
+
## PASO 13 — .env
|
|
608
|
+
|
|
609
|
+
Si existe `.env` y no `.env.example`, crear `.env.example` con nombres de variables sin valores reales.
|
|
610
|
+
Agregar `.env` a `.gitignore` si falta.
|
|
611
|
+
Nunca copiar secretos.
|
|
612
|
+
|
|
613
|
+
## PASO 14 — Validación final
|
|
614
|
+
|
|
615
|
+
Validar:
|
|
616
|
+
|
|
617
|
+
```txt
|
|
618
|
+
AGENTS.md existe
|
|
619
|
+
docs/ai completo según modo
|
|
620
|
+
docs/skills si modo standard/enterprise
|
|
621
|
+
.cursor/rules existe
|
|
622
|
+
.cursor/commands si standard/enterprise
|
|
623
|
+
.cursor/agents si enterprise
|
|
624
|
+
evals si IA
|
|
625
|
+
legacy pointers sin duplicación larga
|
|
626
|
+
.env ignorado
|
|
627
|
+
AGENTS.md no ignorado
|
|
628
|
+
sin contradicciones evidentes
|
|
629
|
+
```
|
|
630
|
+
|
|
631
|
+
## PASO 15 — Reporte final
|
|
632
|
+
|
|
633
|
+
Entregar:
|
|
634
|
+
|
|
635
|
+
```md
|
|
636
|
+
# Harness installation report
|
|
637
|
+
|
|
638
|
+
## Resumen
|
|
639
|
+
|
|
640
|
+
## Modo aplicado
|
|
641
|
+
|
|
642
|
+
## Archivos creados
|
|
643
|
+
|
|
644
|
+
## Archivos modificados
|
|
645
|
+
|
|
646
|
+
## Archivos legacy convertidos en punteros
|
|
647
|
+
|
|
648
|
+
## Fuente de verdad
|
|
649
|
+
|
|
650
|
+
## Stack detectado
|
|
651
|
+
|
|
652
|
+
## Gentle AI
|
|
653
|
+
|
|
654
|
+
## Engram/Graphify
|
|
655
|
+
|
|
656
|
+
## Campos pendientes [COMPLETAR]
|
|
657
|
+
|
|
658
|
+
## Riesgos o excepciones
|
|
659
|
+
|
|
660
|
+
## Validación
|
|
661
|
+
|
|
662
|
+
## Próximos pasos
|
|
663
|
+
```
|
|
664
|
+
|
|
665
|
+
## Regla final
|
|
666
|
+
|
|
667
|
+
No entregues solo recomendaciones.
|
|
668
|
+
|
|
669
|
+
Ejecuta los cambios directamente cuando tengas permisos.
|
|
670
|
+
|
|
671
|
+
Si no puedes escribir archivos, entrega un patch organizado por rutas.
|
|
672
|
+
|
|
673
|
+
## v2 — Universal-first adapters
|
|
674
|
+
|
|
675
|
+
El harness no debe depender de Cursor ni de un modelo específico.
|
|
676
|
+
|
|
677
|
+
Arquitectura obligatoria:
|
|
678
|
+
|
|
679
|
+
```txt
|
|
680
|
+
Core universal:
|
|
681
|
+
- AGENTS.md
|
|
682
|
+
- docs/ai/
|
|
683
|
+
- docs/skills/
|
|
684
|
+
- docs/specs/
|
|
685
|
+
- evals/
|
|
686
|
+
|
|
687
|
+
Adapters:
|
|
688
|
+
- .cursor/
|
|
689
|
+
- .codex/
|
|
690
|
+
- .claude/
|
|
691
|
+
- .github/
|
|
692
|
+
- .gemini/
|
|
693
|
+
- .gentle-ai/
|
|
694
|
+
- CLAUDE.md
|
|
695
|
+
- GEMINI.md
|
|
696
|
+
```
|
|
697
|
+
|
|
698
|
+
Crear en modo standard/enterprise:
|
|
699
|
+
|
|
700
|
+
```txt
|
|
701
|
+
docs/ai/model-policy.md
|
|
702
|
+
docs/ai/provider-routing.md
|
|
703
|
+
docs/ai/tool-adapters.md
|
|
704
|
+
docs/ai/context-budget.md
|
|
705
|
+
docs/skills/model-selection.md
|
|
706
|
+
docs/skills/tool-adapter-sync.md
|
|
707
|
+
```
|
|
708
|
+
|
|
709
|
+
Crear en enterprise:
|
|
710
|
+
|
|
711
|
+
```txt
|
|
712
|
+
.codex/skills/sdd/SKILL.md
|
|
713
|
+
.codex/skills/tdd/SKILL.md
|
|
714
|
+
.codex/skills/evals/SKILL.md
|
|
715
|
+
.codex/skills/checkpoint/SKILL.md
|
|
716
|
+
.claude/agents/
|
|
717
|
+
.claude/skills/
|
|
718
|
+
.gentle-ai/
|
|
719
|
+
.cursor/agents/sdd/
|
|
720
|
+
```
|
|
721
|
+
|
|
722
|
+
Regla:
|
|
723
|
+
|
|
724
|
+
```txt
|
|
725
|
+
Universal core first.
|
|
726
|
+
Tool adapters second.
|
|
727
|
+
Model providers third.
|
|
728
|
+
```
|
|
729
|
+
|
|
730
|
+
DeepSeek/modelos económicos pueden ejecutar tareas verificables y de bajo riesgo.
|
|
731
|
+
Modelos fuertes deben revisar arquitectura, seguridad, specs críticas, eval strategy y aceptación final.
|
|
732
|
+
|
|
733
|
+
## v3 — Loop Engineering + OpenCode adapter
|
|
734
|
+
|
|
735
|
+
Agregar en modo enterprise, y en standard si el usuario usa OpenCode/Gentle AI:
|
|
736
|
+
|
|
737
|
+
```txt
|
|
738
|
+
docs/ai/loops.md
|
|
739
|
+
docs/ai/loop-policy.md
|
|
740
|
+
docs/ai/loop-observability.md
|
|
741
|
+
docs/ai/loop-log.md
|
|
742
|
+
docs/skills/loop-design.md
|
|
743
|
+
docs/skills/loop-debugging.md
|
|
744
|
+
docs/skills/loop-review.md
|
|
745
|
+
docs/skills/loop-retrospective.md
|
|
746
|
+
.opencode/
|
|
747
|
+
.opencode/agents/
|
|
748
|
+
.opencode/loops/
|
|
749
|
+
opencode.json.sample
|
|
750
|
+
.gentle-ai/loops/
|
|
751
|
+
evals/loop-regression/
|
|
752
|
+
```
|
|
753
|
+
|
|
754
|
+
Regla central:
|
|
755
|
+
|
|
756
|
+
```txt
|
|
757
|
+
OpenCode executes.
|
|
758
|
+
Gentle AI structures SDD/TDD.
|
|
759
|
+
DeepSeek iterates cheaply.
|
|
760
|
+
Harness governs.
|
|
761
|
+
Loops repair with boundaries.
|
|
762
|
+
Evals validate.
|
|
763
|
+
Graphify observes dependencies.
|
|
764
|
+
Engram preserves learning.
|
|
765
|
+
Human approves impact.
|
|
766
|
+
```
|
|
767
|
+
|
|
768
|
+
## v4 — Universal Adapter Parity
|
|
769
|
+
|
|
770
|
+
Installers must not make OpenCode the authority.
|
|
771
|
+
|
|
772
|
+
Create/update:
|
|
773
|
+
|
|
774
|
+
```txt
|
|
775
|
+
docs/ai/adapter-parity.md
|
|
776
|
+
docs/ai/governance.md
|
|
777
|
+
docs/skills/adapter-parity-review.md
|
|
778
|
+
.cursor/rules/adapter-parity.mdc
|
|
779
|
+
.cursor/commands/adapter-parity.md
|
|
780
|
+
.cursor/agents/adapter-parity-reviewer.md
|
|
781
|
+
.codex/skills/adapter-parity/SKILL.md
|
|
782
|
+
.codex/skills/loop-engineering/SKILL.md
|
|
783
|
+
.claude/agents/adapter-parity-reviewer.md
|
|
784
|
+
.claude/skills/loop-engineering/SKILL.md
|
|
785
|
+
.opencode/agents/adapter-parity-reviewer.md
|
|
786
|
+
.pi/
|
|
787
|
+
```
|
|
788
|
+
|
|
789
|
+
Rule:
|
|
790
|
+
|
|
791
|
+
```txt
|
|
792
|
+
No adapter is primary by authority.
|
|
793
|
+
An adapter can be primary only by workflow preference.
|
|
794
|
+
The core universal remains governance-primary.
|
|
795
|
+
```
|
|
796
|
+
|
|
797
|
+
OpenCode is workflow-primary only when the user uses OpenCode as runtime.
|
|
798
|
+
Cursor, Codex and Claude must preserve SDD/TDD/evals/loops/checkpoint parity.
|
|
799
|
+
|
|
800
|
+
## v5 — Enforcement-first upgrade
|
|
801
|
+
|
|
802
|
+
Install or update:
|
|
803
|
+
|
|
804
|
+
```txt
|
|
805
|
+
docs/ai/enforcement.md
|
|
806
|
+
docs/ai/quality-gates.md
|
|
807
|
+
docs/ai/eval-strategy.md
|
|
808
|
+
docs/ai/trust-policy.md
|
|
809
|
+
docs/ai/trust-allowlist.md
|
|
810
|
+
docs/ai/installer-cli.md
|
|
811
|
+
docs/ai/observability-runtime.md
|
|
812
|
+
docs/ai/rollback-runtime.md
|
|
813
|
+
docs/ai/maintainability-gates.md
|
|
814
|
+
.github/workflows/harness-quality-gate.yml
|
|
815
|
+
.github/workflows/harness-security-gate.yml
|
|
816
|
+
.github/dependabot.yml
|
|
817
|
+
scripts/harness/
|
|
818
|
+
evals/golden/
|
|
819
|
+
evals/tool-calls/
|
|
820
|
+
evals/schema/
|
|
821
|
+
evals/regression/
|
|
822
|
+
```
|
|
823
|
+
|
|
824
|
+
## v6 — Spec sizing
|
|
825
|
+
|
|
826
|
+
Install spec complexity classification:
|
|
827
|
+
|
|
828
|
+
```txt
|
|
829
|
+
docs/ai/spec-sizing.md
|
|
830
|
+
docs/ai/spec-intake.md
|
|
831
|
+
docs/ai/spec-escalation.md
|
|
832
|
+
docs/specs/templates/basic-spec.md
|
|
833
|
+
docs/specs/templates/standard-spec.md
|
|
834
|
+
docs/specs/templates/complex-spec.md
|
|
835
|
+
docs/skills/spec-complexity-classifier.md
|
|
836
|
+
docs/skills/spec-intake.md
|
|
837
|
+
docs/skills/spec-escalation-review.md
|
|
838
|
+
```
|
|
839
|
+
|
|
840
|
+
Adapters should support:
|
|
841
|
+
|
|
842
|
+
```txt
|
|
843
|
+
/spec-size
|
|
844
|
+
/spec-intake
|
|
845
|
+
/spec-escalate
|
|
846
|
+
```
|
|
847
|
+
|
|
848
|
+
Rule:
|
|
849
|
+
|
|
850
|
+
```txt
|
|
851
|
+
Do not force complex SDD on simple tasks.
|
|
852
|
+
Do not allow basic specs for high-impact work.
|
|
853
|
+
```
|