| name | researching-repos |
| description | Explore repository structure, map dependencies, and surface conventions before making changes. Use when onboarding to a new repo, investigating unfamiliar code, or preparing context for a review.
|
Researching Repos
Systematically explore a repository to build the context needed for effective code review or contribution.
Steps
-
Identify the stack — read root config files (package.json, pyproject.toml, go.mod) to determine languages, frameworks, and dependency versions.
-
Map the structure — walk the top-level directories and identify their roles (source, tests, config, docs, scripts, CI).
-
Find entry points — locate main files, route registrations, CLI entrypoints, or event handlers that anchor the request/data flow.
-
Trace dependencies — follow imports from entry points to understand which modules depend on which, and where shared types or utilities live.
-
Discover conventions — read 2-3 representative source files to learn naming patterns, error handling style, test organization, and commit message format.
-
Note pain points — flag areas with high complexity, outdated dependencies, missing tests, or inconsistent patterns.
Output
Produce a concise summary covering:
- Technology stack and versions
- Directory layout with purpose annotations
- Key entry points and data flow paths
- Established conventions worth preserving
- Areas of concern or technical debt