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-sonnet-opus-haikugit 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-sonnet-opus-haiku)<a href="https://agentmods.dev/skills/mauricioperera/kdd/pm-sonnet-opus-haiku"><img src="https://agentmods.dev/badge/skills/mauricioperera/kdd/pm-sonnet-opus-haiku/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-sonnet-opus-haiku"><img src="https://agentmods.dev/badge/skills/mauricioperera/kdd/pm-sonnet-opus-haiku.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Anti-Refusal · line 67 Skill instructs the agent to omit warnings, disclaimers, or ethical commentary. Stripping safety caveats hides risk from the user and is a common jailbreak preamble.Fix: Remove instructions that suppress warnings, disclaimers, or ethical commentary. Let the agent surface safety-relevant caveats to the user.
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 | $0.00277 | $0.06841 |
| Opus 5.5 | $0.00111 | $0.02736 |
| Sonnet 5.5 | $0.00055 | $0.01368 |
| Haiku 4.5 | $0.00028 | $0.00684 |
Grade A, and why
pm-sonnet-opus-haiku 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 2mo 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 — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sonnet 5 = PM · Opus 4.8 = implementador · Haiku 4.5 = QA
Patrón genérico, independiente del repo. El repo de trabajo es el cwd salvo que el usuario indique otro: sustituí <REPO> por esa ruta. Sirve para una sola tarea o para orquestar varias en paralelo.
Por qué esta asignación (no la inviertas)
- El orquestador quema turnos y contexto durante TODA la sesión; el implementador solo por tarea acotada. Por eso el barato dirige y el caro implementa: Sonnet 5 rinde cerca de Opus 4.8 en trabajo agéntico a fracción del precio, y trae 1M de contexto por defecto, que es exactamente lo que un PM necesita para sostener plan, specs y veredictos sin compactar.
- Opus 4.8 se reserva para donde su capacidad extra rinde: implementación no trivial, arquitectura, bugs finos.
- QA en Haiku 4.5, misma familia que el implementador: se pierde la diversificación de sesgos entre proveedores que daba un modelo externo, a cambio de cero dependencia externa (sin login, sin cuenta cloud, sin tags que se retiran) y de read-only GARANTIZADO por ausencia de tools de escritura en la definición del subagent — no por una bandera de permisos que hay que confiar a ciegas. Si más adelante querés diversidad real de familia, se puede volver a un QA externo (Kimi, GPT, etc.) sin tocar el resto del patrón.
Prerrequisitos
- Sesión principal en Sonnet 5: arrancá con
claude --model claude-sonnet-5o cambiá con/model claude-sonnet-5. Confirmalo antes de empezar; si la sesión corre en otro modelo, avisale al usuario. - Subagent implementador definido en
.claude/agents/implementador-opus.md(plantilla abajo). Si no existe, crealo como primer paso. - Subagent QA definido en
.claude/agents/qa-haiku.md(plantilla abajo). Si no existe, crealo como primer paso, junto con el implementador. - Subagent verificador adversarial definido en
.claude/agents/verificador-adversarial-haiku.md(plantilla abajo). Solo hace falta si vas a usar el paso opcional de verificación adversarial (Capa 3); crealo la primera vez que lo necesites.
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.
- 2mo ago First seen · 186 lines · 277 tokens per session scan A abcab840b828
pm-sonnet-opus-haiku is a skill published in the GitHub repository MauricioPerera/KDD (11 stars, last pushed yesterday), licensed MIT. It adds 277 tokens to every session and 6,841 once invoked, about $0.0011 per session on Opus 5.5. 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
crit-cli
Use when an agent needs to author or reply to crit inline comments programmatically (including multi-agent workflows commenting on shared code/plans/docs/proposals), publish or unpublish a crit review with crit share, sync a crit review to or from a GitHub PR or GitLab MR, or read/interpret a crit review JSON file.…
crit
Review code changes, a plan, a live page (running dev server), or a local HTML file with Crit inline comments and structured human feedback. Use only when the user explicitly invokes /crit or directly asks to use Crit; a generic review request does not count.
crit-story
Author a crit story and continue the interactive review loop only when the user explicitly invokes crit-story or directly asks you to generate a crit story. Do not infer this skill from generic review, PR, or diff-review requests.
docs-governance
Route broad documentation-governance requests to existing ECC skills and run an opt-in, read-only audit of mapped documentation roles, links, ADR indexes, and evidence references. Use when a user asks for documentation governance, project-knowledge organization, documentation integrity, or a deterministic docs audit…
meeting-action-items
Turn meeting notes into cited decisions, owners, tickets.
coordinator
Take over as coordinator of the current LangWatch drive: read the live state from disk (the lane roster, the newest handover, the shape counters), report it, then work the handover's next action by writing manifests and spawning lanes. Replaces pasting a brief out of a handover document - the state is read, never…