スキル一覧に戻る
wayne930242

project-discovery

by wayne930242

A skills-driven Agentic Context Engineering workflow for Claude Code.

7🍴 2📅 2026年1月9日
GitHubで見るManusで実行

SKILL.md


name: project-discovery description: Deep project analysis for architecture planning. Use when starting migration or designing new agent components for an unfamiliar codebase.

Project Discovery

Systematically explore a project to produce a discovery report for agent-architect.

Process

1. Structure Scan

Explore the project root and key directories:

□ Root files: package.json, pyproject.toml, Cargo.toml, go.mod, etc.
□ Source directories: src/, lib/, app/, components/
□ Config files: tsconfig.json, .eslintrc, prettier, etc.
□ CI/CD: .github/workflows/, .gitlab-ci.yml, Jenkinsfile
□ Existing Claude setup: .claude/, CLAUDE.md

2. Tech Stack Identification

Determine:

AspectExamples
Language(s)TypeScript, Python, Rust, Go
FrameworkReact, Next.js, FastAPI, Express
Build toolsVite, Webpack, esbuild, tsc
Package managernpm, pnpm, yarn, pip, cargo
Test frameworkJest, Vitest, pytest, go test

3. Workflow Discovery

Identify existing automation and manual workflows:

Automated:

  • Build scripts in package.json / Makefile
  • CI/CD pipelines
  • Pre-commit hooks (husky, lint-staged)
  • Existing Claude hooks

Manual (candidates for automation):

  • Common developer commands
  • Deployment procedures
  • Release processes

4. Pattern Recognition

Look for:

  • Code conventions (formatting, naming)
  • Architecture patterns (DDD, clean arch, MVC)
  • Testing patterns (unit, integration, e2e)
  • Documentation style

Output: Discovery Report

Produce a structured report:

# Discovery Report: [Project Name]

## Overview
- **Type**: [Web app / CLI / Library / API / Monorepo]
- **Language**: [Primary language(s)]
- **Framework**: [Main framework(s)]

## Tech Stack
| Category | Technology |
|----------|------------|
| Runtime | ... |
| Framework | ... |
| Build | ... |
| Test | ... |
| Lint/Format | ... |

## Existing Automation
- [List current CI/CD, hooks, scripts]

## Recommended Components

### Skills
- [ ] `skill-name` - [reason]

### Commands
- [ ] `/command-name` - [reason]

### Hooks
- [ ] `hook-name` on [event] - [reason]

### Rules
- [ ] `rule-name` - [convention to extract]

## Notes
[Any special considerations, legacy code, migration risks]

Handoff

After producing the report:

  1. Present findings to user for validation
  2. Pass approved recommendations to agent-architect
  3. Let agent-architect classify and delegate each component

Exploration Tips

  • Use Glob for file patterns: **/*.config.*, **/test/**
  • Use Grep for conventions: import, export default, async function
  • Check README.md for documented workflows
  • Look at recent git commits for active development areas

スコア

総合スコア

55/100

リポジトリの品質指標に基づく評価

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

0/10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新

+5
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

+5

レビュー

💬

レビュー機能は近日公開予定です