povvo/claudikins-github-readme-for-perfectionists/deep-dive
Claude Code and agent skill by povvo/claudikins-github-readme-for-perfectionists
- community
Checking that the SKILL.md file is still in the repository.
Use when starting the README pipeline to extract technical facts from codebase. Identifies project type, tech stack, dependencies, entry points, CI configuration. Outputs Reality Report.
Install
npx vskill@latest install povvo/claudikins-github-readme-for-perfectionists/deep-diveSKILL.md
The Deep Dive
Goal: Establish the ground truth of the codebase (what IS) before imagining what could be.
Workflow
Step 0: Load Intelligence Read references/what-to-extract.md for detailed extraction patterns.
- Scout Structure: List files/dirs to understand layout (
bin,src,dist,docs). - Identify Configuration: Read dependency files (
package.json,Cargo.toml,pyproject.toml). - Extract Entry Points: Find main exports, CLI commands, API routes.
- Context Extraction: Scan chat history for user struggles/intent (if available).
- Synthesize: Generate the "Codebase Reality Report".
1. Project Type Detection
Entry Point Analysis
| Files Present | Likely Type | |
|---|---|---|
bin/, shebang in main | CLI tool | |
src/index.ts with exports | Library | |
src/main.ts or app.ts | Application | |
plugin.json, .claude-plugin/ | Plugin | |
src/server.ts, app.py | API/Server | |
Dockerfile only | Container | |
setup.py with console_scripts | Python CLI | |
Cargo.toml with [[bin]] | Rust CLI |
Package.json Indicators
2. Tech Stack Extraction
Dependency Files
| File | Ecosystem | |
|---|---|---|
package.json | Node.js/JavaScript | |
requirements.txt, pyproject.toml | Python | |
Cargo.toml | Rust | |
go.mod | Go | |
Gemfile | Ruby | |
pom.xml, build.gradle | Java | |
composer.json | PHP |
Framework Detection
JavaScript/TypeScript:
react,vue,angular- Frontend frameworkexpress,fastify,koa- Backend frameworkelectron- Desktop appreact-native- Mobile app
Python:
fastapi,flask,django- Web frameworkclick,typer,argparse- CLI frameworkpytorch,tensorflow- ML framework
Build Tool Detection
| Tool | Files | |
|---|---|---|
| TypeScript | tsconfig.json | |
| Webpack | webpack.config.js | |
| Vite | vite.config.ts | |
| esbuild | esbuild in scripts | |
| Rollup | rollup.config.js | |
| Babel | .babelrc, babel.config.js |
3. Dependencies & Prerequisites
Runtime Requirements
From package.json:
From pyproject.toml:
From Cargo.toml:
System Dependencies
Check these files:
Dockerfile- RUN apt-get install commands.github/workflows/*.yml- setup stepsMakefile- prerequisite checksINSTALL.mdordocs/installation.md
Common system deps:
libssl-dev- cryptographylibpq-dev- PostgreSQLffmpeg- media processinggraphviz- diagram generation
4. Entry Points & API Surface
What does the project export/expose?
- Main function/export in index file
- CLI commands from bin directory
- API routes from server files
- Public methods from library exports
CLI help text extraction:
Value proposition sources:
package.jsondescription field- Module-level docstrings
- JSDoc on main exports
- Existing README fragments
5. CI/Automation Detection
GitHub Actions
Badge sources from workflows:
- Workflow status - Build badge
codecov/codecov-action- Coverage badge- Release workflow - Version badge
Other CI
| Service | Config File | |
|---|---|---|
| CircleCI | .circleci/config.yml | |
| Travis | .travis.yml | |
| GitLab CI | .gitlab-ci.yml | |
| Jenkins | Jenkinsfile |
6. Chat History Context
Source: ~/.claude/history.jsonl
Filter entries where project matches the current project path.
Pattern Extraction
Repeated questions (knowledge gaps):
- "How do I..." patterns
- "Why isn't..." patterns
- "What's the difference between..." patterns
Problem categories:
- Configuration/Setup
- Debugging/Errors
- Feature implementation
- Performance
- Testing
- Deployment
README Implications
| Pattern Found | README Implication | |
|---|---|---|
| Repeated config questions | Detailed configuration section | |
| Debugging struggles | Troubleshooting section | |
| "How do I test" questions | Testing section with examples | |
| Performance questions | Performance section | |
| Deployment questions | Deployment guides |
Output Format
Important
- Focus on FACTS, not improvements (that's crystal-ball's job)
- Be specific with file:line references
- Note confidence levels:
- High: Extracted from config files
- Medium: Inferred from patterns
- Low: Guessed from structure
Handoff
After presenting the Reality Report:
- Ask: "Does this accurately represent the codebase? Any missing information we need to clarify?"
- Wait for user confirmation.
- Transition:
- If changes needed: Refine the Deep Dive.
- If approved: "Proceeding to crystal-ball to predict what this project COULD become."
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/deep-dive/SKILL.md
Popularity
- Repository stars
- 3
- Repository forks
- 1
- 7-day trend score
- 7
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/povvo/claudikins-github-readme-for-perfectionists/deep-dive)