setup-fellowship

Explore repository structure and generate a tailored CLAUDE.md for project calibration.

Updated Feb 10, 2026
One-click install
npx skills add https://github.com/endaoment/fellowship-code --skill setup-fellowship
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: setup-fellowship
Source: https://github.com/endaoment/fellowship-code/tree/main/skills/setup-fellowship
Command: npx skills add https://github.com/endaoment/fellowship-code --skill setup-fellowship

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Calibrate the entire Fellowship to a specific project by exploring the codebase and asking targeted follow-ups, generating a tailored CLAUDE.md and configuring each agent for the project.

Core Features & Use Cases

  • Explore the repository to identify project identity, structure, tech stacks, and commands.
  • Auto-detect workflows, agent roles, and conventions to craft a project-specific CLAUDE.md.
  • Calibrate agent knowledge and roster, updating essential docs and guidelines to align with the codebase.

Quick Start

Run the /setup-fellowship command in your editor to begin calibrating Gandalf for your project.

Frequently Asked Questions about setup-fellowship

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate codebase exploration to generate a CLAUDE.md file for my project?

Automate codebase calibration by analyzing repository structure, manifests, and CI configs to generate a tailored CLAUDE.md. The tool identifies project identity, tech stacks, and commands to create project-specific agent knowledge templates.

How does repository calibration work for monorepo projects with multiple stacks?

Repository calibration works by exploring single-repo or monorepo structures to identify individual repos, stacks, and workflows. It then generates agent-specific domain knowledge templates and updates essential docs to align with the detected codebase conventions.

Can I use this setup tool to configure AI agent roles for both frontend and backend operations?

Yes, the setup tool configures AI agent roles across frontend, backend, and operations. It auto-detects workflows and conventions to calibrate agent knowledge and roster, applying tailored domain knowledge templates for each specific stack.

What is the best way to configure an AI agent to understand my existing project conventions?

The best way to configure an AI agent is through automated calibration phases that explore the codebase, report structural gaps, and generate a project-specific CLAUDE.md. This aligns the agent's knowledge base with your root manifests and CI configurations.

Do I need specific dependencies to calibrate my repository structure for an AI agent?

No specific dependencies are required to calibrate your repository structure. The calibration process directly analyzes your existing root manifests and CI configs to produce the tailored agent knowledge templates without external setup.

What are the limitations of using setup-fellowship for codebase calibration?

The setup-fellowship calibration is limited to analyzing repository structure, manifests, and CI configs. It focuses on generating CLAUDE.md and agent knowledge templates, relying on existing codebase conventions rather than modifying the underlying source code.