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 delegar-poolgit 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/delegar-pool)<a href="https://agentmods.dev/skills/mauricioperera/kdd/delegar-pool"><img src="https://agentmods.dev/badge/skills/mauricioperera/kdd/delegar-pool/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/delegar-pool"><img src="https://agentmods.dev/badge/skills/mauricioperera/kdd/delegar-pool.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 | $0.00147 | $0.01590 |
| Opus 5 | $0.00073 | $0.00795 |
| Sonnet 5 | $0.00029 | $0.00318 |
| Haiku 4.5 | $0.00015 | $0.00159 |
Grade A, and why
delegar-pool 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 10d 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 — 100 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Delegar a un dev efímero vía pool exec (headless) — mecánica agnóstica a la tarea
Patrón de lanzamiento reutilizable para pool (poolside.ai), análogo en rol a
delegar-ollama pero para el CLI pool en vez de
ollama launch claude. Esta skill NO sabe de CCDD ni de qué tarea estás
delegando — solo de cómo invocar pool exec sin perder tiempo ni confiar
ciegamente en su reporte. Verificado en vivo 2026-07-26 con contratos KDD
reales en Python y Rust.
Rol (misma doctrina que delegar-glm-ccdd)
pool es un implementador efímero, no un par de confianza. Yo (el
orquestador) redacto el contrato + tests congelados, delego la implementación,
y verifico el resultado yo mismo — nunca confío en el exit code ni en el
resumen que pool imprime al final.
Comando base
pool exec --unsafe-auto-allow -d <REPO_O_DIR> -f <prompt.txt> -o json > <log.txt> 2>&1
--unsafe-auto-allowOBLIGATORIO en no-interactivo. Sin él, la primera acción que requiera aprobación (leer O escribir, no importa cuál) falla con"approval required in non-interactive mode: encountered unexpected error"y exit code 1. No existe modo headless sin este flag — no está en el README, pero sí enpool exec --help.- Prompt en ARCHIVO (
-f), nunca inline (-p) para instrucciones multilínea — mismo motivo que endelegar-ollama(comillas/escaping). -o jsonda NLJSON estructurado (reasoning/thought/toolCall/toolCallResult/assistantMessage) — más auditable que texto plano, no hace falta parsear logs crudos.-d <dir>fija el cwd de la sesión; no asumas que hereda el cwd del shell.- Exit codes:
0= tarea completa,4= fallo explícito reportado porpool, cualquier otro = error inesperado (incluida la falta de--unsafe-auto-allow).
El riesgo real NO es que se cuelgue esperando aprobación
Verificado dos veces (lectura primero, escritura primero) SIN el flag: el
rechazo de aprobación es casi instantáneo (segundos), con error explícito, no
un cuelgue. El riesgo real es el tiempo de razonamiento ANTES de llegar a
esa acción. Trayectoria real auditada
(%LOCALAPPDATA%\poolside\trajectories\trajectory-standalone_*.ndjson):
27m30s de corrida total, de los cuales ~27 minutos fueron tres bloques de
thought consecutivos sin ninguna tool call, y el rechazo de aprobación en sí
tardó 24ms. Conclusión: lanzá SIEMPRE con --unsafe-auto-allow desde el
arranque en un directorio desechable — no hay forma de "probar primero sin el
flag" barato, porque el costo no está en el rechazo sino en el razonamiento
previo, que podés terminar pagando igual.
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.
- 10d ago First seen · 100 lines · 147 tokens per session scan A a557e49dc755
delegar-pool is a skill published in the GitHub repository MauricioPerera/KDD (7 stars, last pushed yesterday), licensed MIT. It adds 147 tokens to every session and 1,590 once invoked, about $0.0007 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
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.
trio-omnigent
Run the Cursor-backed Omnigent Trio loop from the current Claude/Codex UI session when the user explicitly says “Trio Omnigent”, “Omnigent Trio”, or invokes /trio-omnigent. Do not use for an ordinary native Trio request.
trio-bridge
Suggest background Trio work by scanning a project and writing deduplicated proposal cards into proposals/.
research-a-evolve
Provides guidance for automatically evolving and optimizing AI agents across any domain using LLM-driven evolution algorithms. Use when building self-improving agents, optimizing agent prompts and ...