agentleFS
dominionthedevdominionthedev/skills/codebase-radar/SKILL.md

codebase-radar

Skill1 starsChanged 6 months agoMITView on GitHub

Instantly scans and understands any codebase — producing an architecture map, tech stack detection, complexity hotspots, dependency graph, and entry points — so Claude can work on the codebase as if it's already familiar with it. Use this skill whenever the user shares a codebase, asks Claude to work on their project, says "here's my repo", "look at this codebase", "help me with my project", "fix something in my code", "add a feature to", or provides a path to an existing project. ALWAYS run the analyze script FIRST before reading any individual files — it gives a 10x better understanding in 1 second than reading files one by one. Also use when debugging, refactoring, or reviewing a repo.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Start with entry points

    line 62
  2. Distinctive:

    Read hotspots that are relevant to the task

    line 63

≈ “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

Codebase Radar — Instant Repo Intelligence

You understand any codebase immediately by running the radar script first. This is always your first move when working with an existing project.


The Golden Rule

Never start reading files one by one. Run the radar first.

The analyze script gives you tech stack, architecture map, hotspots, entry points, and dependencies in under a second. That context makes every subsequent file read 10x more meaningful.


Workflow

Step 1 — Run the analyze script

python skills/codebase-radar/scripts/analyze.py <path> [--depth 6]
# Examples
python analyze.py /path/to/project
python analyze.py . --depth 4
python analyze.py ~/projects/my-app --json

Step 2 — Parse the JSON output

Key fields to extract:

FieldUse it for
summary.frameworksKnow the stack instantly
summary.primary_languageSet your language lens
architectureUnderstand the directory roles
entry_pointsFind where execution starts
hotspotsKnow which files need the most care
dependenciesUnderstand what libraries are in play

Step 3 — Targeted file reading

Now that you have the map, read only the relevant files:

  • Start with entry points
  • Read hotspots that are relevant to the task
  • Navigate architecture dirs by their role

Step 4 — Begin the task

Armed with full context, complete the user's request efficiently.


Output Reference

{
  "project": "my-app",
  "summary": {
    "total_files": 142,
    "total_lines": 18420,
    "primary_language": "TypeScript",
    "frameworks": ["Next.js", "Tailwind", "Prisma", "PostgreSQL"]
  },
  "languages": {
    "TypeScript": { "files": 87, "lines": 12400 },
    "CSS": { "files": 12, "lines": 980 }
  },
  "architecture": {
    "app": { "files": 34, "role": "app/routes" },
    "components": { "files": 28, "role": "UI components" },
    "lib": { "files": 15, "role": "shared utilities" }
  },
  "entry_points": ["src/app/layout.tsx", "src/app/page.tsx"],
  "hotspots": [
    { "file": "src/lib/auth.ts", "lines": 480, "functions": 22 }
  ],
  "dependencies": {
    "package.json": { "prod": ["next", "react", "prisma"] }
  }
}

What the script detects

Frameworks (signals-based detection): Next.js, Astro, React, Vue, Svelte, Flutter, FastAPI, Django, Go Chi/Gin/Fiber, Axum, Tokio, Prisma, Drizzle, PostgreSQL, MongoDB, Redis, Docker, GitHub Actions, Tailwind, shadcn/ui, and more.

Languages: Python, Go, Rust, TypeScript, JavaScript, Dart, Swift, Kotlin, Java, C++, C, C#, Ruby, PHP, Zig, Elixir, SQL, GraphQL, Protobuf

Hotspots: Files over 200 lines with many function definitions — these are the complex areas that need most attention during refactoring or debugging.


Script location

skills/codebase-radar/scripts/analyze.py

Requires Python 3.10+ stdlib only. Runs in < 1 second on most codebases.

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.