seeding-a-project

Generate CLAUDE.md, AGENTS.md, and .agents configuration files for repositories.

18|9|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/arozumenko/sdlc-skills --skill seeding-a-project
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: seeding-a-project
Source: https://github.com/arozumenko/sdlc-skills/tree/main/skills/seeding-a-project
Command: npx skills add https://github.com/arozumenko/sdlc-skills --skill seeding-a-project

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It eliminates the manual, error-prone setup work required to prepare a new repository for role-based agent collaboration and workflow execution.

Core Features & Use Cases

  • Generates project context files: creates CLAUDE.md, AGENTS.md, and a structured .agents/ directory so agents can work with consistent, project-specific information.
  • Captures team workflow signals: runs a PR-sampling survey to infer how the team actually operates (roles, review gates, branching/commit conventions, test delivery patterns) and writes them into the generated .agents documents.
  • Documents project systems for downstream skills: captures issue tracker, test management system, knowledge base, bug filing style, and automation PR policy for runtime by other skills.
  • Supports evidence-based tool wiring: (as part of later steps) can generate .agents/team-comms.md and wire host-specific tool permissions for restrictive IDEs/hosts.

Quick Start

Use the seeding-a-project skill during onboarding or when the user asks to seed the project, so your repo ends up with CLAUDE.md, AGENTS.md, and a populated .agents/ configuration set.

Frequently Asked Questions about seeding-a-project

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

FAQPage Schema
How do I set up AI agent configuration files for a new repository?

AI agent configuration files for a new repository are generated by creating CLAUDE.md, AGENTS.md, and a structured .agents/ directory using templates that capture project-specific information for role-based agent collaboration and workflow execution.

What is repository onboarding for AI-assisted software delivery?

Repository onboarding for AI-assisted software delivery is the process of seeding a project with agent-ready configs, which eliminates manual setup work by capturing team workflow signals and documenting project systems for downstream agents.

How does PR sampling infer team workflows during project seeding?

PR sampling infers team workflows by surveying the pull request history to extract how the team actually operates, including roles, review gates, branching conventions, and test delivery patterns, then writing them into generated .agents documents.

Do I need project-root write access to seed agent-ready project configs?

Yes, project-root write access is required to seed agent-ready project configs because the process produces CLAUDE.md, AGENTS.md, and .agents/* documents directly in the repository structure.

What project systems are documented for downstream AI agents during repo onboarding?

Repo onboarding documents issue trackers, test management systems, knowledge bases, bug filing styles, and automation PR policies so downstream agents can reference them at runtime.

Can I wire MCP tool permissions for restrictive IDEs after seeding a project?

Yes, MCP tool permissions for restrictive IDEs can be wired after seeding a project, as the process can generate .agents/team-comms.md and configure host-specific tool permissions as part of later steps.