Repository Exploration
- 1 installs
- Updated June 3, 2026
- adriebaselmans/shooter
repository-exploration is a Claude Code skill that analyzes an existing repository and produces a compact reusable brief of its architecture and conventions.
About
repository-exploration is a Claude Code skill that quickly analyzes an existing codebase and captures the architecture and conventions that matter. A developer or another agent role uses it to get grounded repository context without rescanning the whole repo. It writes a compact brief and a facts file into a shared knowledge store.
- Analyzes a repository quickly and extracts architecture and conventions
- Produces a compact reusable brief.md and facts.yaml
- Separates direct evidence from informed inference
Repository Exploration by the numbers
- 1 all-time installs (skills.sh)
- Ranked #1,361 of 1,879 Documentation skills by installs in the Skillselion catalog
- Data as of Jul 7, 2026 (Skillselion catalog sync)
repository-exploration capabilities & compatibility
- Capabilities
- repository exploration · codebase analysis · architecture documentation
- Use cases
- documentation · research
- Pricing
- Free
What repository-exploration says it does
Analyze an existing code repository quickly, extract the architecture and conventions that matter, and create a compact reusable repository brief for the AI team or the user.
Produce compact reusable knowledge instead of repeating the same discovery work.
npx skills add https://github.com/adriebaselmans/shooter --skill repository-explorationAdd your badge
Show developers this skill is listed on Skillselion. Paste this into your README.
| Installs | 1 |
|---|---|
| Last updated | June 3, 2026 |
| Repository | adriebaselmans/shooter ↗ |
What it does
Analyze a repository fast and produce a compact reusable brief of its architecture and conventions.
Who is it for?
Teams or agents that need grounded repository context before starting work on a codebase.
Skip if: Deep line-by-line audits; it explicitly avoids rewriting the whole repository into memory.
When should I use this skill?
When the user asks to base work on a repository, requests a repo analysis, or another role is blocked on missing context.
What you get
A compact, reusable repository brief and facts file another role can use without rescanning the repo.
- brief.md, facts.yaml and an updated _index.yaml in the repository knowledge store
By the numbers
- 7-step procedure
- 3 required output files (brief.md, facts.yaml, _index.yaml)
Files
Repository Exploration
Use this skill when the user asks to base work on a specific repository, asks for a repo analysis directly, or when another team role is blocked on missing repository context.
Goals
- Build grounded understanding of the target repository fast.
- Produce compact reusable knowledge instead of repeating the same discovery work.
- Surface what is known, what is inferred, and what remains uncertain.
Required Inputs
- Target repository path, URL, or checkout location
- The user request or downstream role question that motivates the exploration
- Existing repository knowledge in
.ai-team/framework/memory/wiki/repositories/when available
Required Outputs
.ai-team/framework/memory/wiki/repositories/<repo-slug>/brief.md.ai-team/framework/memory/wiki/repositories/<repo-slug>/facts.yaml- Updated
wiki/repositories/_index.yamlwhen the repository is new to the knowledge store
Procedure
1. Identify the target repository, its local path or source URL, and its current revision when available. 2. Detect the stack from manifests, lockfiles, build files, and top-level docs. 3. Find the important entry points, packages, services, and runtime boundaries. 4. Trace the user-relevant or task-relevant flows instead of reading the entire repo indiscriminately. 5. Record key modules, architectural patterns, naming conventions, testing approach, and build or run commands. 6. Separate direct evidence from informed inference. 7. Hand off compact findings that another role can use without rescanning the same repo.
Exploration Rules
- Prefer breadth-first orientation first, then drill into task-relevant areas.
- Use file references and revision identifiers so later agents can verify claims quickly.
- Keep the brief short enough to fit comfortably into future context windows.
- Capture open questions explicitly when the evidence is incomplete.
- Do not rewrite the whole repository into memory.
Brief Shape
The brief should cover:
- Repository identity and revision
- Purpose and top-level architecture
- Stack and tooling
- Key directories and entry points
- Important flows or subsystems for the current task
- Conventions, risks, and extension points
- Open questions
name: repository-exploration
description: Analyze a repository quickly and produce compact grounded repository knowledge.
response_mode: one-shot
required_inputs:
- target repository path, URL, or checkout location
- user request or downstream role question
- existing repository knowledge in .ai-team/framework/memory/repository-knowledge/ when available
expected_outputs:
- .ai-team/framework/memory/repository-knowledge/<repo-slug>/brief.md
- .ai-team/framework/memory/repository-knowledge/<repo-slug>/facts.json
- .ai-team/framework/memory/repository-knowledge/index.md when the repository is new
owned_artifacts:
- .ai-team/framework/memory/repository-knowledge/<repo-slug>/brief.md
- .ai-team/framework/memory/repository-knowledge/<repo-slug>/facts.json
completion_criteria:
- The brief captures structure, conventions, risks, and open questions.
- Evidence paths and provenance are included in the output.
Related skills
Lark MarkdownInstantly turn any markdown file into clean, formatted Lark/Figma-compatible documents without manual reformatting.488k
Lark DocCreate, read, update, summarize, rewrite and manage Feishu/Lark cloud documents directly from agent workflows.401k16.1kLark WikiCreate, organize, query, and manage documents, spaces, and members inside Lark (Feishu) knowledge bases directly from AI agent workflows.397k16.1k
Domain ModelingBuild and sharpen a project's domain model: pin down domain terminology in a glossary and record architectural decisions as ADRs.318k202k
Opensource Guide CoachGet expert, attribution-safe guidance on launching and sustaining an open-source project.270k71Readme I18nMaintain consistent, updatable language selector blocks across every README variant in a multilingual repository.270k71