What is _composicion?

An internal protocol for dynamically assembling Alfred's development team based on the task, technology stack, and runtime signals. It also defines default autonomy settings for a project.

Use it when running Alfred workflows such as feature work, fixes, audits, experiments, or releases, especially when configuring project autonomy and team composition.

It gives Alfred's commands a shared process for deciding how the team should work and what level of independence each area has.

_composicion — Command by 686f6c61

Install

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.

agentmods

npx agentmods add commands/686f6c61/alfred-dev/_composicion

Clone the repo

git clone --depth 1 https://github.com/686f6c61/alfred-dev

Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 2,017 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00022	$0.02017
Opus 5	$0.00011	$0.01009
Sonnet 5	$0.00004	$0.00403
Haiku 4.5	$0.00002	$0.00202

Measured 2d ago against content hash 172fa1d893cb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

_composicion 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.

commands/_composicion.md · 195 lines

How it starts

The opening of the file, as written. The whole thing — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Protocolo de composición dinámica de equipo

Este fichero define el protocolo compartido para componer el equipo de cada sesión. Lo usan todos los comandos de Alfred (feature, quick, fix, spike, audit, ship). Cualquier cambio aquí se refleja en todos los flujos.

Paso 0 -- Configuración inicial del proyecto

Antes de cualquier otra cosa, comprueba si el proyecto ya tiene configurado el modo de autonomía. Lee .claude/alfred-dev.local.md y busca la sección autonomia: en el frontmatter YAML.

Si la sección autonomia: NO existe (primera vez que se usa Alfred en este proyecto):

Escribe directamente una configuración por defecto compatible con Claude Code CLI en .claude/alfred-dev.local.md:

autonomia:
  producto: autonomo
  arquitectura: autonomo
  desarrollo: autonomo
  calidad: autonomo
  documentacion: autonomo
  entrega: autonomo

NO uses AskUserQuestion en este bootstrap inicial. El objetivo es que Alfred pueda actuar automáticamente desde la primera sesión si el usuario invoca /alfred, /alfred-dev:feature, /alfred-dev:quick, /alfred-dev:fix, /alfred-dev:spike, /alfred-dev:audit o /alfred-dev:ship.
Muestra un mensaje breve indicando que Alfred ha activado el modo autopilot por defecto para evitar bloquear el flujo en la primera sesión y que el usuario puede cambiarlo más tarde con /alfred-dev:ajustes.

Si la sección autonomia: YA existe: salta este paso y continúa directamente.

Nota: el usuario puede cambiar el modo en cualquier momento con /alfred-dev:ajustes.

Paso 1 -- Contexto del proyecto

Llama a suggest_optional_agents(project_dir) para obtener señales basadas en I/O del proyecto (stack detectado, presencia de ORM, frontend, HTML público, remote GitHub, tamaño del proyecto, memoria activa). Estas señales son objetivas y complementan tu razonamiento semántico.

Paso 2 -- Razonamiento semántico

Lee la descripción de la tarea y las señales del proyecto. Decide qué agentes opcionales son relevantes usando tu comprensión semántica, no keywords. Razona sobre el dominio de la tarea, no sobre palabras sueltas.

Read the full file on GitHub · 195 lines

Changes

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 · 195 lines · 22 tokens per session scan A 172fa1d893cb

Subscribe to this mod's changes

_composicion is a command published in the GitHub repository 686f6c61/alfred-dev (119 stars, last pushed 18d ago), licensed MIT. It adds 22 tokens to every session and 2,017 once invoked, about $0.0001 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 2026-08-30.

Building agents Planning ai-agents automation claude-code claude-code-plugin sdlc

cdd-quick-create

Produce a small, self-contained deliverable — a script plus a README, not a full CDD project: /cdd-quick-create (takes an optional one-line description of what to build).

drabaioli/cdd · 0 tokens

auto-build

Fully autonomous product builder. Takes a PRD/spec and builds the entire product without human intervention. Usage: /auto-build or /auto-build 'product description' --max-iterations 100 [--workflow for background, resumable Workflows orchestration].

alphaaiservice/cortex · 58 tokens

init-mcp-server

Scaffold a standalone MCP (Model Context Protocol) server project. Supports Python (official mcp SDK) and TypeScript (@modelcontextprotocol/sdk). Use this for pure MCP servers (Slack-MCP, Postgres-MCP, GitHub-MCP, etc.) — for MCP servers embedded inside a SaaS app, use /auto-build or /retrofit instead. Usage…

alphaaiservice/cortex · 121 tokens

feature-map

Build a visual feature dependency map of the codebase with Mermaid diagrams. Usage: /feature-map or /feature-map payments.

alphaaiservice/cortex · 24 tokens

init-claude-plugin

Scaffold a new Claude Code plugin (commands, agents, skills, hooks, MCP servers, validator, GitHub Actions). Dogfoods every plugin best practice Cortex itself has learned through v1.0–v1.2. Usage: /init-claude-plugin [--with-commands] [--with-agents] [--with-skills] [--with-hooks] [--with-mcp]…

alphaaiservice/cortex · 97 tokens

claude-flow-swarm

Coordinate multi-agent swarms for complex tasks.

ruvnet/RuView · 14 tokens