agentleFS
Baba01hacker666Baba01hacker666/dummyskills/skills/repo-inspector/SKILL.md

repo-inspector

Skill0 starsChanged 2 months agoMITView on GitHub

Safely inspects an unfamiliar codebase to map its tech stack, architecture, entry points, build/test commands, and key directories.

  • Reads credentials

The file

Show its rules and text

We split this skill into the rules it gives an agent

18 rules. About 0 are common, 7 are distinctive, and 11 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 11

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill when:

    line 14
  2. Very specific:

    Do not use to run deep code review or diff analysis.

    line 20
  3. Very specific:

    Do not use to plan multi-step feature implementations.

    line 21
  4. Very specific:

    Do not use to diagnose a specific broken command.

    line 22
  5. Distinctive:

    If no workspace root is specified, use the current working directory.

    line 28
  6. Very specific:

    If the directory does not exist or is empty, stop and report: "Error: Directory <path> is empty or does not exist."

    line 29
  7. Distinctive:

    Do not run commands in parent directories outside the workspace root.

    line 30
  8. Distinctive:

    Follow these numbered actions:

    line 33
  9. Very specific:

    Check directory listing at top level.

    line 35
  10. Distinctive:

    Read the manifest file to identify runtime version, dependencies, scripts, and build tools.

    line 37
  11. Distinctive:

    Do not execute package installation or build scripts during inspection.

    line 38
  12. Very specific:

    Search for main.*, index.*, src/index.*, cmd/, app/, or CLI binaries declared in manifest.

    line 40
  13. Very specific:

    Check test script in manifest (npm test, pytest, cargo test, go test ./...).

    line 42
  14. Very specific:

    Verify that .env files, .git/, and secret credentials are NOT read into public output.

    line 45
  15. Very specific:

    If multiple manifests exist (e.g.

    line 50
  16. Very specific:

    If no manifest is found: inspect file extensions (.py, .sh, .c, .rs) to determine language.

    line 51
  17. Very specific:

    If documentation files exist (README.md, CONTRIBUTING.md): check quickstart command lines.

    line 52
  18. Distinctive:

    Stop and ask the user when:

    line 93

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Repository Inspector

Safely maps an unfamiliar repository. Discovers the project structure, language, package manager, entry points, and test commands without modifying files.

1. Purpose

Provide a quick, accurate map of a repository so subsequent development skills know where to locate source code, configurations, and test runners.

2. When to Use

Use this skill when:

  • Arriving in a new or unfamiliar codebase.
  • User asks "What does this repo do?" or "How do I run this project?".
  • Pre-flight step before diagnosing bugs or applying code changes.

3. When Not to Use

  • Do not use to run deep code review or diff analysis; use diff-reviewer.
  • Do not use to plan multi-step feature implementations; use task-brief.
  • Do not use to diagnose a specific broken command; use command-failure-diagnoser.

4. Required Inputs

  • workspace_root: The file path to the repository root directory (default: current working directory).

5. Missing-Input Behavior

  1. If no workspace root is specified, use the current working directory.
  2. If the directory does not exist or is empty, stop and report: "Error: Directory <path> is empty or does not exist."
  3. Do not run commands in parent directories outside the workspace root.

6. Procedure

Follow these numbered actions:

  1. Action 1: List Root Files and Directories
    • Check directory listing at top level. Look for manifest files: package.json, Cargo.toml, pyproject.toml, go.mod, pom.xml, Makefile.
  2. Action 2: Identify Languages and Package Manager
    • Read the manifest file to identify runtime version, dependencies, scripts, and build tools.
    • Do not execute package installation or build scripts during inspection.
  3. Action 3: Discover Key Entry Points
    • Search for main.*, index.*, src/index.*, cmd/, app/, or CLI binaries declared in manifest.
  4. Action 4: Discover Test Setup
    • Check test script in manifest (npm test, pytest, cargo test, go test ./...).
    • Identify test directory (tests/, __tests__/, spec/, *_test.go).
  5. Action 5: Safety and Secret Boundary Check
    • Verify that .env files, .git/, and secret credentials are NOT read into public output.
  6. Action 6: Emit Repository Map
    • Format the findings using Section 8.

7. Decision Rules

  • If multiple manifests exist (e.g. monorepo with root package.json and nested packages): record repository type as Monorepo and list top-level workspaces.
  • If no manifest is found: inspect file extensions (.py, .sh, .c, .rs) to determine language.
  • If documentation files exist (README.md, CONTRIBUTING.md): check quickstart command lines; do not read entire large files into context.

8. Output Format

Emit the repository overview using this structured template:

### Repository Map: [Project Name or Folder]

- **Primary Language / Stack**: [e.g., TypeScript / Node.js 22]
- **Package Manager / Build Tool**: [e.g., pnpm, npm, cargo, pip]
- **Architecture Style**: [Single package | Monorepo | CLI | Library | Web App]

#### Key Directories & Files
- `src/` - [Core source files]
- `tests/` - [Test suites]
- `config/` - [Configuration files]
- Entry point: `[path to entry file]`

#### Standard Commands
- **Install**: `[e.g., npm install]`
- **Build**: `[e.g., npm run build]`
- **Test**: `[e.g., npm test]`
- **Lint / Check**: `[e.g., npm run lint]`

#### Observations & Warnings
- [Notable findings, missing test configs, or special setup requirements]

9. Verification Checklist

  • Root manifest was inspected without executing unverified code.
  • Primary language, test command, and build command are verified from files.
  • Entry point exists and was confirmed on disk.
  • No secrets or personal data were captured in the output.

10. Common Failures and Recovery

Failure SymptomCauseRecovery Action
Unrecognized stackProprietary or missing manifestCheck file extension counts and look for shell scripts or Makefiles.
Monorepo confusionManifests in multiple subdirectoriesInspect pnpm-workspace.yaml, lerna.json, or root workspaces field.
Slow or huge treeWalking node_modules or .gitAlways ignore .git, node_modules, dist, target, venv.

11. Stop or Escalation Conditions

Stop and ask the user when:

  • The workspace contains no recognisable source code or build configuration.
  • The directory contains encrypted files or requires proprietary credentials to inspect.

12. Examples

Example 1: Successful Execution

Input:

Inspect workspace: /workspace/sample-service
Files found: package.json, tsconfig.json, src/server.ts, src/routes.ts, tests/server.test.ts, README.md
package.json scripts: { "build": "tsc", "test": "vitest run", "start": "node dist/server.js" }

Expected Observable Actions:

  1. Inspected package.json scripts and dependencies.
  2. Verified entry point at src/server.ts.
  3. Verified test runner is vitest run.

Output:

### Repository Map: sample-service

- **Primary Language / Stack**: TypeScript / Node.js
- **Package Manager / Build Tool**: npm / tsc
- **Architecture Style**: Web Service (Backend API)

#### Key Directories & Files
- `src/` - Application source code
- `src/server.ts` - Main HTTP server entry point
- `tests/` - Automated test suites
- `tsconfig.json` - TypeScript compiler configuration

#### Standard Commands
- **Install**: `npm install`
- **Build**: `npm run build`
- **Test**: `npm test`
- **Start**: `npm start`

#### Observations & Warnings
- Test command uses Vitest (`vitest run`).
- Output builds into `dist/`.

Example 2: Ambiguous or Failure Case

Input:

Inspect workspace: /workspace/empty-project (0 files present)

Output:

### Repository Map: Error - Empty Repository

- **Primary Language / Stack**: Unknown
- **Status**: Halted

The directory `/workspace/empty-project` is completely empty. No source files, manifests, or build scripts were found. Please clone the project repository or specify a populated workspace folder.

Example 3: Expected Final Output

Output:

### Repository Map: auth-gateway

- **Primary Language / Stack**: Python 3.11 / FastAPI
- **Package Manager / Build Tool**: Poetry (`pyproject.toml`)
- **Architecture Style**: REST Microservice

#### Key Directories & Files
- `auth_gateway/` - Main Python package
- `auth_gateway/main.py` - FastAPI app initialization
- `tests/` - Unit and integration tests using pytest
- `pyproject.toml` - Dependencies and tool settings

#### Standard Commands
- **Install**: `poetry install`
- **Build**: `poetry build`
- **Test**: `poetry run pytest`
- **Lint / Check**: `poetry run ruff check .`

#### Observations & Warnings
- Dockerfile present for container deployment.
- No local `.env` required if environment variables are provided.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.