Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add MauricioPerera/KDD --skill pm-native-ccddgit clone --depth 1 https://github.com/MauricioPerera/KDDWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mauricioperera/kdd/pm-native-ccdd)<a href="https://agentmods.dev/skills/mauricioperera/kdd/pm-native-ccdd"><img src="https://agentmods.dev/badge/skills/mauricioperera/kdd/pm-native-ccdd/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mauricioperera/kdd/pm-native-ccdd"><img src="https://agentmods.dev/badge/skills/mauricioperera/kdd/pm-native-ccdd.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00254 | $0.02506 |
| Opus 5 | $0.00127 | $0.01253 |
| Sonnet 5 | $0.00051 | $0.00501 |
| Haiku 4.5 | $0.00025 | $0.00251 |
Grade A, and why
pm-native-ccdd scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PM nativo · Devs = sub-agentes de la app (Agent + model override) · QA = CCDD gate
Misma metodología CCDD que pm-glm-ccdd — contrato + tests congelados
autorados ANTES de delegar, verify-by-artifact, política de reintentos, red-team del HECHO,
suite 2x. Lo único que cambia es el mecanismo de delegación: en vez de ollama launch
(GLM externo), el implementador es un sub-agente nativo vía la tool Agent.
Tiering (los cuatro tiers, nativos)
- Triage → Haiku 4.5 (
Agent,model:"haiku"): leer issue, clasificar, limpiar log, mapear tools MCP, detectar si ya está hecho. Cortá el pipeline aquí si no hay trabajo. - PM/oráculo → Sonnet 5 (
model:"sonnet") por defecto; Opus solo para oráculos adversariales sin tropo conocido. Autora contrato + tests congelados + lint + baseline rojo. - Dev/implementador → Haiku 4.5 (
model:"haiku"), escalá amodel:"sonnet"una función puntual si Haiku falla el gate 2×. - Gate/verify → el orquestador (vos): corrés
mcp__ccdd-complexity__*+ re-corrés los tests.
Cómo invocar cada agente (native)
Triage / PM / Dev — tool Agent:
Agent({ subagent_type: "claude", model: "haiku"|"sonnet"|"opus",
run_in_background: true, isolation: "worktree" (solo devs en paralelo),
description: "<3-5 palabras>", prompt: "<spec autocontenida>" })
- El resultado (texto final del agente) vuelve como tool result; el harness reporta
subagent_tokens/tool_uses/duration_msen la notificación → medición de costo exacta. run_in_background: truepara no bloquear (llega notificación al terminar).
Paralelismo sin colisión — isolation: "worktree":
- Cada dev en su propio git worktree → NO necesitás declarar "archivos disjuntos" (lo resuelve el aislamiento). Auto-limpiado si no cambió. Coste ~200-500ms + disco por agente: úsalo SOLO cuando varios devs mutan archivos a la vez; para tareas secuenciales, omitilo.
Reintento tras FAIL — SendMessage al MISMO dev (conserva contexto):
SendMessage({ to: "<agentId>", summary: "gate FAIL: cyclomatic 12>10",
... "extraé helpers, no toques los tests, re-corré node --test" })
- Más barato que un
Agentnuevo: el dev ya tiene el contexto de la tarea; solo le pasás el feedback del gate. (En pm-glm-ccdd cada retry re-lanzaba GLM re-enviando toda la spec.)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 114 lines · 254 tokens per session scan A 8c82cb4bddec
pm-native-ccdd is a skill published in the GitHub repository MauricioPerera/KDD (8 stars, last pushed 3d ago), licensed MIT. It adds 254 tokens to every session and 2,506 once invoked, about $0.0013 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
kotlin-testing
Kotlin testing patterns with Kotest, MockK, coroutine testing, property-based testing, and Kover coverage. Follows TDD methodology with idiomatic Kotlin practices.
tdd-workflow
Use this skill when writing new features, fixing bugs, or refactoring code. Enforces test-driven development with 80%+ coverage including unit, integration, and E2E tests.
react-testing
React component testing with React Testing Library, Vitest/Jest, MSW for network mocking, accessibility assertions with axe, and the decision boundary between component tests and Playwright/Cypress end-to-end runs. Use when writing or fixing tests for React components, hooks, or pages.
python-testing
Python testing best practices using pytest including fixtures, parametrization, mocking, coverage analysis, async testing, and test organization. Use when writing or improving Python tests.
react-patterns
React 18/19 patterns including hooks discipline, server/client component boundaries, Suspense + error boundaries, form actions, data fetching, state management decision trees, and accessibility-first composition. Use when writing or reviewing React components.
rust-testing
Rust testing patterns including unit tests, integration tests, async testing, property-based testing, mocking, and coverage. Follows TDD methodology.