agentleFS
MrCipherSmithMrCipherSmith/goodai-base/skills/autodoc-scanner/SKILL.md

autodoc-scanner

Skill0 starsChanged 5 months agoMITView on GitHub

Phase 1 subagent for autodoc-orchestrator. Scans project structure, detects stack, identifies module boundaries, entry points, and dependencies. Use when: dispatched by autodoc-orchestrator Phase 1. NOT for: direct user invocation.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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:

    Read top-level directory structure.

    line 67

≈ “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

autodoc-scanner

Purpose

Produce a complete structural map of the project that downstream analysts and writers can rely on without re-scanning the codebase.

Iron Laws

#Law
1Scan file tree — do NOT read full file contents (use directory listings and targeted reads of package.json, tsconfig, Makefile, etc.)
2Detect module boundaries from directory structure and build config, not from code semantics
3Always list entry points (main files, index files, app bootstraps)
4Return module list in next_phase_hints.modules[] — orchestrator uses this to spawn analysts

Input Contract

PROJECT_DIR: "<absolute path>"
CONFIG:
  scope: "full | backend | frontend | <custom path>"
  language: "en | ru | both"
  existing_docs: "<path or null>"

Output Contract

status: "DONE" | "NEEDS_CONTEXT"
summary: "<3-5 sentences: project type, module count, stack summary>"
next_phase_hints:
  modules:
    - slug: "backend"
      path: "<absolute path>"
      type: "backend | frontend | shared | mobile | infra | other"
      language: "TypeScript | Python | Go | ..."
      framework: "NestJS | Express | React | ..."
      entry_points: ["<file>"]
    - slug: "frontend"
      path: "<absolute path>"
      ...
  has_api: true | false
  has_schemas: true | false
  has_tests: true | false
artifact_path: "jobs/<job>/artifacts/project-map.md"

Workflow

Step 1: Project Tree Scan

Read top-level directory structure. Identify:

  • Build/package files: package.json, pyproject.toml, go.mod, Cargo.toml, pom.xml, Makefile
  • Workspace configs: nx.json, turbo.json, lerna.json, pnpm-workspace.yaml
  • Entry points: main.*, index.*, app.*, server.*, cmd/
  • Config files: tsconfig.json, vite.config.*, webpack.config.*, docker-compose.yml
  • Existing documentation: README.md, docs/, CONTRIBUTING.md

Step 2: Module Boundary Detection

From the directory structure and workspace config, identify module boundaries:

Monorepo (nx/turbo/lerna) → each package/app is a module
Single repo → detect by: apps/, packages/, src/modules/, src/features/
Backend indicators: api/, server/, backend/, services/
Frontend indicators: web/, client/, frontend/, ui/
Shared/lib indicators: lib/, shared/, common/, packages/

For each detected module, read its package.json or equivalent to extract:

  • Name, description
  • Dependencies (framework detection)
  • Scripts (build, test, dev commands)

Step 3: Stack Detection

For each module, determine:

SignalDetected
"next" in depsNext.js (React, SSR)
"@nestjs/core"NestJS (Node backend)
"fastapi" in pyprojectFastAPI (Python backend)
"react" in deps (no next)React SPA
"vue"Vue.js
go.mod existsGo
Cargo.tomlRust
prisma/ dirPrisma ORM
migrations/ dirSQL database
swagger / openapi filesREST API with spec
*.proto filesgRPC
graphql/ or *.graphqlGraphQL

Step 4: Entry Points

For each module, find entry point files:

  • Node: main.ts, index.ts, app.ts, server.ts
  • Python: main.py, app.py, __main__.py
  • Go: cmd/*/main.go, main.go
  • Frontend: main.tsx, _app.tsx, App.tsx, index.html

Step 5: API + Schema Detection

has_api = true IF:
  - `swagger.json` / `openapi.yaml` found
  - `*.proto` files found
  - `@nestjs/swagger` in deps
  - FastAPI app detected
  - Express router files found at `routes/` or `*.routes.ts`

has_schemas = true IF:
  - `prisma/schema.prisma` found
  - `migrations/` directory found
  - `models/` directory with *.py or *.ts files
  - `*.graphql` schema files found

Step 6: Write project-map.md

# Project Map: <Project Name>

## Overview
- **Type**: monorepo | single-repo
- **Modules**: <N>
- **Languages**: <list>
- **Has API**: yes | no
- **Has DB schemas**: yes | no

## Modules

### <Module Name> (`<path>`)
- **Type**: backend | frontend | shared
- **Language**: <lang>
- **Framework**: <framework + version>
- **Entry point**: `<file>`
- **Key dependencies**: <list>
- **Build command**: `<cmd>`
- **Dev command**: `<cmd>`

## Project Structure (condensed)
<top-2-level directory tree>

## Existing Documentation
<list of found docs, or "none">

## Detected Integrations
<databases, queues, external APIs, auth providers>

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.