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 agentmods add commands/polmarza/project-template/mcp-setupgit clone --depth 1 https://github.com/polmarza/project-templateWhat 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 | $0.00000 | $0.01051 |
| Opus 5 | $0.00000 | $0.00526 |
| Sonnet 5 | $0.00000 | $0.00210 |
| Haiku 4.5 | $0.00000 | $0.00105 |
Grade A, and why
mcp-setup 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 2d 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Configura los servidores MCP de este proyecto según el stack definido en docs/architecture.md.
Sigue el "Protocolo de MCPs" de CLAUDE.md. Este comando lo lanza a demanda; el flujo normal es
que se ejecute solo al terminar la arquitectura o al añadir una integración nueva.
1. Contexto
- Lee
docs/architecture.md. Si el stack todavía no está definido, dilo y para aquí: sin stack no hay nada que decidir. - Ejecuta
claude mcp listpara ver qué servidores hay ya disponibles y en qué estado están.
2. Averigua qué existe
Por cada servicio del stack (base de datos, emails, pagos, despliegue, diseño, observabilidad, gestión de tareas…), determina si publica un servidor MCP.
Si no estás seguro de que exista, de cómo se llama el paquete, de qué transporte usa o de qué
credenciales pide, búscalo en la documentación oficial del servicio antes de proponerlo. No
inventes comandos ni nombres de variables de entorno: un claude mcp add mal copiado deja un
servidor que no arranca y cuesta más depurarlo que buscarlo bien a la primera.
Como fuente vale el dominio del proveedor o su repositorio oficial, nada más. Un blog, un
agregador de MCPs o un gist no son fuente válida para un comando que se va a ejecutar en la
máquina del usuario: un paquete typosquatteado se ejecuta con npx exactamente igual que el
legítimo. Si solo lo encuentras en fuentes no oficiales, dilo y que decida el usuario.
Descarta los servicios sin MCP sin darles vueltas.
3. Pregunta
Presenta la lista de candidatos y, por cada uno, pregunta con qué alcance lo quiere:
| Alcance | Dónde vive | Quién lo ve | Cuándo usarlo |
|---|---|---|---|
Global (user) |
~/.claude.json |
Solo el usuario, en todos sus proyectos | Ya lo tiene configurado o lo usa en todas partes. No se toca nada del repo |
Proyecto (project) |
.mcp.json, commiteado |
Todo el equipo | Recomendado: el servidor forma parte del proyecto y el equipo lo hereda |
Local (local) |
~/.claude.json, bajo la ruta del proyecto |
Solo el usuario, solo aquí | Pruebas o credenciales que no quiere ni referenciadas en el repo |
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.
- 2d ago First seen · 83 lines · 0 tokens per session scan A 286bfa7756ab
mcp-setup is a command published in the GitHub repository polmarza/project-template (79 stars, last pushed 14d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,051 tokens. 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 2026-08-30.
Other commands, from other repositories
claude-flow-swarm
Coordinate multi-agent swarms for complex tasks.
advisor
Advisory gate for triage or plan decisions. Spawns a second-opinion agent that challenges assumptions, surfaces risks, and proposes alternatives before the decision commits. Based on Anthropic advisor tool pattern.
bootstrap
PACT session-start ritual — identify the session team (platform-provisioned), secretary spawn, paused-state surface, bootstrap marker.
ox-session-review
Command "ox-session-review" from sageox/ox, covering failure-mode watch-list (read first), from the ledger root. should print 0, phase 1 — scan & score (read-only), quality buckets (first match wins) and removal candidates.
add-agent
引导新增一个 Agent 适配器。用法 /add-agent.
migrate
Command "migrate" from chohra-med/expo_boilerplate, covering invocation, steps and rule.