| name | repo-onboarding |
| display_name | Repo Onboarding |
| description | Summarizes a repository for humans or AI agents by mapping architecture, commands, conventions, key files, workflows, and safe first tasks. |
| category | documentation |
| version | 1.0.0 |
| tags | ["onboarding","documentation","architecture","repository","agents"] |
| providers | ["claude-code","codex","copilot","openclaw","hermes-agent"] |
| author | agent-skills contributors |
Repo Onboarding
Use this skill when a user or agent needs to quickly understand an unfamiliar repository.
Inputs
- Repository root, file tree, README, package manifests, config files, tests, CI workflows, and recent changes
- Intended audience: contributor, maintainer, reviewer, or AI agent
Steps
- Inspect the top-level structure and identify the main applications, packages, services, and docs.
- Read README and project metadata for stated purpose, setup, and commands.
- Identify build, test, lint, dev server, release, and deployment commands.
- Map important modules, boundaries, data flow, and integration points.
- Note coding conventions, generated files, ignored paths, and risky areas.
- Identify safe first tasks and tasks that require extra context.
- Produce an onboarding summary that can be reused by people or agents.
Output
Return:
- Repository purpose
- Architecture map
- Key files and directories
- Common commands
- Development workflow
- Testing and release notes
- Risks, unknowns, and safe first tasks
Guidelines
- Do not read every file when structural files are enough.
- Distinguish documented facts from inferred architecture.
- Include exact command names when available.
