Fellowship Fellowship Docs

Getting Started

Install Fellowship, initialize your project, and run your first AI-powered task in under 2 minutes.

What is Fellowship?

Fellowship is a CLI that turns AI coding agents into a team that learns. It orchestrates the full cycle: spec → code → review → learn → improve — so each run makes the next one better.

Works with Claude Code, Codex, and any CLI-based AI coding agent.

Why Fellowship? AI coding agents are powerful but stateless. Every session starts from zero — they forget past decisions, repeat mistakes, and ignore project conventions. Fellowship fixes this with persistent memory, auto-generated specs, and built-in code review.

Installation

Install Fellowship globally via npm:

terminal
npm install -g fellowship-cli

Requirements

  • Node.js ≥ 18
  • Git installed and configured
  • A supported AI coding agent CLI (claude or codex)

Run fellowship doctor after installation to verify everything is set up correctly.

New Project

Start a brand new project with Fellowship from scratch:

terminal
# Create project directory
mkdir my-api && cd my-api && git init

# Initialize Fellowship
fellowship init

The init wizard will analyze your description, generate architecture, create an AI agent team, and scaffold the project. It creates:

  • .fellowship/config.yaml — Project config (provider, model, settings)
  • .fellowship/team.yaml — Agent definitions (roles, skills, profiles)
  • .fellowship/profiles/ — Agent profiles with coding guidelines
  • .fellowship/architecture.md — Project architecture reference
  • .fellowship/fellowship.db — SQLite database (runs, decisions, learnings)

Existing Project

Add Fellowship to a project that already has code:

terminal
cd my-existing-project

# Analyze codebase and set up Fellowship
fellowship adopt

The adopt command detects your stack, framework, and patterns, then generates a team tailored to your project.

Running Your First Task

Once initialized, run a task with a natural language description:

terminal
# Run a task
fellowship run "add user authentication with JWT"

# Run in background with human review
fellowship run "add rate limiting" --bg --review
# → You keep working → 🔊 notification → fellowship review
How it works: Fellowship generates a detailed spec from your task description, assigns an agent, executes the work in an isolated git worktree (your working directory never changes), runs AI code review, and extracts learnings for future runs.

Verify Your Setup

Run the doctor command to check everything is working:

terminal
fellowship doctor

⚔️  Fellowship Doctor

  ✅ Node.js v25.6.1  (>= 18)
  ✅ Git 2.47.0
  ✅ Agent CLI  (claude)
  ✅ .fellowship/ directory
  ✅ config.yaml  (anthropic / claude-sonnet-4-6)
  ✅ team.yaml  (4 agents)
  ✅ API key  (configured for anthropic)
  ✅ Sentinel reviewer  (profile present)
  ✅ fellowship.db
  ✅ .gitignore  (covers fellowship.db)
  ⚠️  Learnings  (none yet — run some tasks first)
  ⚠️  Gotchas  (none yet)
  ✅ architecture.md

  All checks passed (11 ✅, 2 ⚠️, 0 ❌)

Next Steps