Outfitter

Outfitter builds effective agent profiles and launches them through wrapped agent CLIs like pi, Claude Code, and future adapters. Individuals, teams, and organizations can share and compose repeatable agent profiles.

If you have not tried Pi yet, Outfitter is the quickest path to a recommended Pi loadout for engineering work.

Quick start

The first time outfitter runs it will start the setup flow to choose profiles and setup a model if pi is not already configured with one.

Run without installing.

npx @ai-outfitter/outfitter

Full install

npm install -g @ai-outfitter/outfitter
outfitter

Outfitter launches agent CLIs; install the agents you plan to use separately.

For the full walkthrough, see Getting started.

Profiles

Profiles compose the context, tools, prompts, skills, extensions, subagents, and DeepWork workflows that shape an agent. Profiles can be shared using Profile Catalog Repos.

# ~/.outfitter/profiles/home-default.yml
id: home-default
label: Home Default
description: Reusable personal defaults for Outfitter-managed Pi runs.
controls:
  provider: openai-codex
  model: gpt-5.5
  thinking: high
  append_system_prompt:
    - |
      Use concise, evidence-backed engineering prose.
      Prefer small, reviewable changes.
      Keep durable decisions in repo files.
    - repo_file: docs/architecture.md

Documentation

• Getting started
• What works today: adapter support matrix
• Profiles
• Profile repositories
• State persistence
• First-time CLI agent users
• Switching to Outfitter
• Documentation index

Use cases:

• Organization profile catalog — Publish shared team roles so new users can start with organization-approved defaults.
• Engineering profile catalog — Package coding, platform, and review profiles for repeatable engineering workflows.
• Persona reviews — Create customer personas to get feedback on ideas, documentation, and designs.

For local development, repository structure, and release workflow details, see Contributing.