Prompt My Project (PMP)

Transform your codebase into AI-ready prompts
Prompt My Project (PMP) is a powerful command-line tool that analyzes your source code
and generates structured prompts for AI assistants like ChatGPT, Claude, and Gemini.

Quick Start • Installation • Features • Usage • Examples

---

🚀 Quick Start

# Install PMP
go install github.com/benoitpetit/prompt-my-project@latest

# Generate a prompt for your current project
pmp prompt .

# Analyze a GitHub repository
pmp github prompt https://github.com/username/project

# Generate a dependency graph
pmp graph . --format stdout:dot | dot -Tpng > graph.png

# Optional: compress context for large projects
pmp prompt . --summary-only           # Architecture overview
pmp prompt . --focus-changes          # Git-aware filtering

---

🎯 Why PMP?

The Problem

When working with AI assistants, you need to share your codebase context efficiently. Manually copying files is tedious and error-prone.

The Solution

PMP automatically generates structured prompts from your codebase:

• 🤖 AI-Ready Prompts: Generate formatted prompts optimized for ChatGPT, Claude, Gemini, and other LLMs
• 📊 Project Analysis: Technology detection, dependencies, code quality metrics
• 🛡️ Security Scanning: Detect secrets, vulnerabilities, dangerous permissions
• 🐙 GitHub Integration: Analyze any repository directly by URL
• 📈 Dependency Graphs: Visual representations of your project structure

---

📦 Installation

Option 1: Go Install (Recommended)

go install github.com/benoitpetit/prompt-my-project@latest

Option 2: Installation Script

Linux/macOS:

curl -fsSL https://raw.githubusercontent.com/benoitpetit/prompt-my-project/master/scripts/install.sh | bash

Windows (PowerShell):

iwr -useb https://raw.githubusercontent.com/benoitpetit/prompt-my-project/master/scripts/install.ps1 | iex

Option 3: Download Binary

Download the latest binary from the releases page

Verify Installation

pmp --help

---

✨ Core Features

🤖 AI-Ready Prompts

Generate structured prompts optimized for ChatGPT, Claude, Gemini, and other LLMs. Output in TXT, JSON, or XML formats.

📊 Visual Dependency Graphs

Create beautiful dependency graphs in multiple formats (DOT, JSON, XML, TXT) for documentation and analysis.

🔍 Project Analysis

• Technology Detection: Automatically identifies 50+ languages, frameworks, and tools
• Dependency Analysis: Multi-language support (Node.js, Go, Python, Rust, Ruby, PHP)
• Code Quality Metrics: Complexity analysis, maintainability index, code smell detection

🛡️ Security Scanning

• Secret Detection: Find exposed API keys, tokens, passwords (AWS, GitHub, JWT, SSH keys)
• Vulnerability Scanning: Built-in CVE database for known vulnerable dependencies
• Permission Analysis: Detect dangerous file permissions and configurations
• Configuration Security: Check Docker, CORS, Nginx configs for security issues

🐙 GitHub Integration

Analyze any GitHub repository directly without manual cloning. Supports HTTPS/SSH URLs and branch selection.

⚡ High Performance

• Parallel processing with configurable workers
• Intelligent file filtering and size limits
• Streaming support for massive projects (handles multi-GB codebases)
• Optimized sorting algorithms (50-90% performance improvement)

🔧 Smart Filtering

• Automatically excludes binary files
• Respects .gitignore rules
• Custom include/exclude patterns with glob support
• File size filtering (min/max)

---

📚 Usage Guide

Basic Commands

Command	Description	Example
pmp prompt	Generate AI-ready prompts	pmp prompt .
pmp github prompt	Analyze GitHub repo	pmp github prompt https://github.com/user/repo
pmp graph	Generate dependency graphs	pmp graph .
pmp github graph	GitHub repo graph	pmp github graph https://github.com/user/repo
pmp completion	Shell completions	pmp completion bash

Generate Prompts

Basic Usage

# Current directory
pmp prompt .

# Specific project
pmp prompt /path/to/project

# Output to stdout (for piping)
pmp prompt . --format stdout:txt

Advanced Filtering

# Include specific file types
pmp prompt . --include "*.go" --include "*.md"

# Exclude patterns
pmp prompt . --exclude "test/*" --exclude "*.log"

# Combine include/exclude
pmp prompt . --include "*.go" --exclude "*_test.go"

# Ignore .gitignore
pmp prompt . --no-gitignore

Size and Performance Controls

# Limit file sizes
pmp prompt . --min-size 100B --max-size 1MB

# Limit total processing
pmp prompt . --max-files 100 --max-total-size 5MB

# Adjust workers (default: number of CPUs)
pmp prompt . --workers 8

Output Formats

# JSON output
pmp prompt . --format json

# XML output
pmp prompt . --format xml

# Stdout for piping
pmp prompt . --format stdout:json | jq .

# Custom output directory
pmp prompt . --output /custom/path

Generate Dependency Graphs

# DOT format (default)
pmp graph .

# JSON structure
pmp graph . --format json

# Text tree
pmp graph . --format txt

# Create visualization
pmp graph . --format stdout:dot | dot -Tpng > graph.png
pmp graph . --format stdout:dot | dot -Tsvg > graph.svg
pmp graph . --format stdout:dot | dot -Tpdf > graph.pdf

GitHub Repository Analysis

# Basic analysis
pmp github prompt https://github.com/username/repo

# Specific branch
pmp github prompt https://github.com/username/repo --branch develop

# Generate graph
pmp github graph https://github.com/username/repo --format dot

# With filtering
pmp github prompt https://github.com/username/repo --include "*.go" --exclude "*test*"

# Multiple formats
pmp github prompt https://github.com/username/repo --format json
pmp github graph https://github.com/username/repo --format stdout:dot | dot -Tpng > graph.png

Supported URL Formats:

• https://github.com/user/repo
• https://github.com/user/repo.git
• git@github.com:user/repo.git

Features:

• Shallow cloning for speed (depth=1)
• Automatic cleanup of temporary files
• Full support for all PMP features
• Works with private repos (via SSH keys or credentials)

Context Compression

For large projects, PMP provides optional flags to compress the context:

AST-Based Summarization (--summary-only)

Generate only function signatures and interfaces instead of full source code:

# Generate architecture overview
pmp prompt . --summary-only

# Combine with filters
pmp prompt . --summary-only --include "*.go" --max-files 100

Supported Languages: Go, JavaScript/TypeScript, Python, Java, C/C++

Git-Aware Context (--focus-changes)

Prioritize recently modified files:

# Focus on modified files
pmp prompt . --focus-changes

# Include last N commits
pmp prompt . --focus-changes --recent-commits 5

Custom Patterns (--summary-patterns)

Manually specify files to summarize:

# Summarize vendor/dependencies
pmp prompt . --summary-patterns "vendor/**" --summary-patterns "node_modules/**"

When to use:

• Very large projects (>100 files)
• Architecture documentation needs
• Pull request focused analysis
• Active development workflows

---

🔬 Examples

For AI Assistants

# Quick prompt for ChatGPT/Claude
pmp prompt . --format stdout:txt | pbcopy                      # macOS
pmp prompt . --format stdout:txt | xclip -selection clipboard  # Linux

# Structured JSON for API integration
pmp prompt . --format stdout:json | curl -X POST https://api.example.com/analyze -d @-

Code Analysis Workflows

# Analyze only source code (no tests, no docs)
pmp prompt . --include "*.go" --include "*.js" --include "*.py" \
  --exclude "*test*" --exclude "*.md"

# Large project with limits
pmp prompt . --max-files 200 --max-total-size 15MB --workers 8

# Get project statistics
pmp prompt . --format stdout:json | jq '.statistics'

# List detected technologies
pmp prompt . --format stdout:json | jq '.technologies[]'

Documentation Generation

# Project structure
pmp graph . --format txt > STRUCTURE.md

# Visual dependency graph
pmp graph . --format stdout:dot | dot -Tsvg > dependencies.svg

# Multiple documentation formats
pmp graph . --format json > structure.json
pmp graph . --format xml > structure.xml

Integration Examples

With jq (JSON processing)

# Get project statistics
pmp prompt . --format stdout:json | jq '.statistics'

# List detected technologies
pmp prompt . --format stdout:json | jq '.technologies[]'

# List key files
pmp prompt . --format stdout:json | jq '.key_files[]'

# Get file types distribution
pmp prompt . --format stdout:json | jq '.file_types[]'

# Extract file paths and sizes
pmp prompt . --format stdout:json | jq '.files[] | {path: .path, size: .size}'

# Count files by extension
pmp prompt . --format stdout:json | jq '.file_types[] | "\(.extension): \(.count)"'

CI/CD Integration

# Generate project context for automated reviews
pmp prompt . --format json --output build/context.json

# Check minimum file count (fail if too few files)
pmp prompt . --format stdout:json | jq -e '.statistics.file_count > 0'

# Verify technologies detected
pmp prompt . --format stdout:json | jq -e '.technologies | length > 0'

# Create documentation automatically
pmp graph . --format txt --output docs/STRUCTURE.md

External Repository Analysis

# Audit open source project
pmp github prompt https://github.com/vendor/project --format stdout:json | jq '.statistics'

# Compare branches
pmp github prompt https://github.com/user/repo --branch main --format stdout:json > main.json
pmp github prompt https://github.com/user/repo --branch dev --format stdout:json > dev.json
diff <(jq -S . main.json) <(jq -S . dev.json)

# List technologies in external library
pmp github prompt https://github.com/vendor/library --format stdout:json | jq '.technologies[]'

# Visual graph of external project
pmp github graph https://github.com/user/project --format stdout:dot | dot -Tsvg > graph.svg

# Get project statistics
pmp github prompt https://github.com/user/project --format stdout:json | jq '.statistics'

---

🛡️ Security Features

PMP includes comprehensive security scanning for your codebase.

Secret Detection

Detects 13 types of secrets in your code:

• AWS Credentials: Access Keys, Secret Keys
• GitHub Tokens: Personal Access Tokens, OAuth tokens
• JWT Tokens: JSON Web Tokens
• SSH Keys: Private keys in code
• API Keys: Generic API keys, Stripe, Slack, etc.
• OAuth Secrets: Client secrets
• Database URLs: Connection strings with credentials
• Passwords: Hard-coded passwords
• Webhooks: Slack webhooks and similar

File Permission Analysis

• World-writable files (chmod 777)
• Executable configuration files
• Private keys without proper permissions (should be 600)
• Sensitive files with insecure permissions

Configuration Security

• Dockerfile: Missing USER directive, exposed SSH ports, running as root
• Nginx: Missing security headers, insecure configurations
• CORS: Wildcard misuse, overly permissive settings
• Environment Files: Exposed .env files in repository

Vulnerability Scanning

Built-in CVE database for detecting vulnerable dependencies:

• Node.js: lodash, axios, express, and more
• Python: django, flask, requests, and more
• Ruby: rails, devise, and more
• PHP: symfony, laravel components, and more

Dangerous Files Detection

• .env files in repository
• Private key files (.pem, .key, id_rsa)
• Credential files
• Database dumps
• Configuration files with secrets

Usage Examples

# Analyze codebase and output to file
pmp prompt . --format txt

# Get project overview in JSON
pmp prompt . --format stdout:json | jq '.'

# Check which technologies are detected
pmp prompt . --format stdout:json | jq '.technologies[]'

# List all key files in the project
pmp prompt . --format stdout:json | jq '.key_files[]'

# Get file count and size statistics
pmp prompt . --format stdout:json | jq '.statistics'

# Note: Security scanning, dependency analysis, and code quality metrics
# are performed during analysis but detailed results are shown in TXT format.
# For full analysis, use: pmp prompt . --format txt

---

🎓 Advanced Features

Technology Detection

Automatically detects 50+ technologies with confidence scoring:

Languages:

• Go, JavaScript, TypeScript, Python, Java, Ruby, PHP
• C#, C, C++, Rust, Kotlin, Swift, Scala
• R, Dart, Lua, Perl, Elixir, Haskell

Frameworks:

• React, Vue, Angular, Svelte
• Django, Flask, FastAPI
• Rails, Sinatra
• Laravel, Symfony
• Express, Koa, NestJS
• Spring, Spring Boot

Build Tools:

• Make, CMake, Gradle, Maven
• npm, yarn, pnpm
• pip, poetry, pipenv
• cargo, bundle, composer

Databases:

• PostgreSQL, MySQL, MongoDB, Redis
• SQLite, Elasticsearch, Cassandra

Testing:

• Jest, Mocha, Pytest, RSpec, PHPUnit
• JUnit, TestNG, Go testing

CI/CD:

• GitHub Actions, GitLab CI, Jenkins
• Travis CI, CircleCI

Containers:

• Docker, Docker Compose, Kubernetes

Dependency Analysis

Multi-language dependency analysis with detailed insights:

Supported Ecosystems:

• Node.js: package.json, package-lock.json, yarn.lock, pnpm-lock.yaml
• Go: go.mod, go.sum
• Python: requirements.txt, Pipfile, Pipfile.lock, pyproject.toml, poetry.lock
• Rust: Cargo.toml, Cargo.lock
• Ruby: Gemfile, Gemfile.lock
• PHP: composer.json, composer.lock

Features:

• Dependency classification (production, dev, optional)
• Version information and constraints
• Vulnerability detection via CVE database
• Obsolete package detection
• Dependency count and metrics

Usage:

# View dependency analysis (shown in TXT format)
pmp prompt . --format txt

# Export full analysis including dependencies
pmp prompt . --format json

# Note: Dependency details are analyzed and shown in the TXT output.
# The JSON format includes basic project information and file contents.

Code Quality Metrics

Comprehensive code quality analysis:

Metrics Provided:

• Cyclomatic Complexity: Per-file complexity analysis
• Maintainability Index: 0-100 score based on complexity, code size, and structure
• Code Smells: Automatic detection of common anti-patterns
• Test Coverage Estimate: Based on test file patterns

Code Smells Detected:

• Long functions (>50 lines)
• Long files (>500 lines)
• High complexity functions
• Deeply nested code
• Too many parameters
• Duplicate code patterns

Usage:

# View quality metrics (shown in TXT format)
pmp prompt . --format txt

# Export full analysis including quality metrics
pmp prompt . --format json

# Note: Code quality analysis is performed and details are shown in TXT output.
# Use the TXT format to see complexity, maintainability, and code smells.

Linter Integration

Integrates with popular linters when available:

Supported Linters:

• JavaScript/TypeScript: ESLint
• Go: golint, staticcheck, go vet
• Python: pylint
• Ruby: RuboCop

Features:

• Automatic linter detection
• Formatted issue summary
• Severity classification (error, warning, info)
• Fall back to basic linting when external tools unavailable

Usage:

# View linter results (shown in TXT format)
pmp prompt . --format txt

# Note: Linter integration output is included in the TXT format.
# External linters are automatically detected and run when available.

Streaming Support

For massive projects, PMP automatically uses streaming mode:

Features:

• Memory-efficient processing for >100MB projects
• Configurable chunk sizes
• Progress tracking
• Handles projects of any size
• Automatic detection and activation

When Streaming Activates:

• Projects with >100MB total size
• More than 10,000 files
• Available system memory below threshold
• Manual activation via configuration

---

⚙️ Configuration

Environment Variables

# Output directory
export PMP_OUTPUT_DIR="./analysis"

# Worker count
export PMP_WORKERS=8

# Output format
export PMP_FORMAT="json"

# File limits
export PMP_MAX_FILES=500
export PMP_MAX_TOTAL_SIZE="10MB"
export PMP_MIN_SIZE="1KB"
export PMP_MAX_SIZE="100MB"

# Patterns
export PMP_EXCLUDE="vendor/**,node_modules/**"
export PMP_INCLUDE="*.go,*.js,*.py"

Project Configuration (.pmprc)

Create .pmprc in your project root:

{
  "exclude": ["vendor/**", "node_modules/**", "dist/**", "build/**", ".git/**"],
  "include": ["*.go", "*.js", "*.ts", "*.py", "*.java", "*.md"],
  "minSize": "1KB",
  "maxSize": "100MB",
  "maxFiles": 500,
  "maxTotalSize": "10MB",
  "format": "txt",
  "outputDir": "pmp_output",
  "workers": 8,
  "noGitignore": false,
  "summaryOnly": false,
  "focusChanges": false,
  "recentCommits": 3,
  "summaryPatterns": [
    "vendor/**",
    "node_modules/**",
    "**/generated/**",
    "**/*.pb.go"
  ]
}

Shell Autocompletion

Enable autocompletion for your shell:

Bash:

source <(pmp completion bash)

# For all sessions (Linux):
pmp completion bash > /etc/bash_completion.d/pmp

# For all sessions (macOS):
pmp completion bash > /usr/local/etc/bash_completion.d/pmp

Zsh:

echo 'autoload -U compinit; compinit' >> ~/.zshrc
pmp completion zsh > "${fpath[1]}/_pmp"

Fish:

pmp completion fish | source

# For all sessions:
pmp completion fish > ~/.config/fish/completions/pmp.fish

PowerShell:

pmp completion powershell | Out-String | Invoke-Expression

# Add to $PROFILE for all sessions

---

🔧 Troubleshooting

Permission Denied

# Make binary executable
chmod +x pmp

# Or install via go install
go install github.com/benoitpetit/prompt-my-project@latest

Command Not Found

# Check if Go bin is in PATH
echo $PATH | grep -q "$(go env GOPATH)/bin"

# If not, add to PATH
echo 'export PATH="$PATH:$(go env GOPATH)/bin"' >> ~/.bashrc
source ~/.bashrc

Large Project Performance

# Limit processing
pmp prompt . --max-files 200 --max-total-size 15MB

# Adjust workers
pmp prompt . --workers 4

# Use specific patterns
pmp prompt . --include "*.go" --include "*.js"

Output Too Large

# Focus on specific files
pmp prompt . --include "*.go" --exclude "*test*" --max-files 50

# Size limits
pmp prompt . --max-size 50KB --max-total-size 5MB

# Optional: use context compression
pmp prompt . --summary-only                          # Architecture overview
pmp prompt . --focus-changes                         # Git-aware filtering
pmp prompt . --summary-patterns "vendor/**"          # Custom patterns

Missing Small Files

# Include very small files
pmp prompt . --min-size 0

Git Context Not Working

# Ensure you're in a git repository
git status

# If not initialized, initialize git
git init
git add .
git commit -m "Initial commit"

# Now use focus-changes
pmp prompt . --focus-changes

Binary Detection Issues

# Force include files if wrongly detected as binary
pmp prompt . --no-gitignore --min-size 0

# Or add to .pmprc:
{
  "noGitignore": true,
  "minSize": "0"
}

---

📊 Project Statistics

View comprehensive project statistics:

# Project statistics
pmp prompt . --format stdout:json | jq '.statistics'

# Technology stack
pmp prompt . --format stdout:json | jq '.technologies'

# File types distribution
pmp prompt . --format stdout:json | jq '.file_types'

# Key files
pmp prompt . --format stdout:json | jq '.key_files'

# Potential issues
pmp prompt . --format stdout:json | jq '.issues'

# Get complete analysis (includes security, dependencies, quality)
pmp prompt . --format txt

# Export everything for processing
pmp prompt . --format json

---

🤝 Contributing

Contributions are welcome! Here's how you can help:

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

Development Setup

# Clone the repository
git clone https://github.com/benoitpetit/prompt-my-project.git
cd prompt-my-project

# Build
go build -o pmp .

# Run tests
go test ./...

# Run benchmarks
go test -bench=. ./pkg/analyzer/

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

---

💖 Support

If you find PMP useful, consider supporting its development:

• ⭐ Star the repository
• 🐛 Report bugs and request features via Issues
• 💰 Donate on Liberapay
• 📢 Share with your network

---

🔗 Links

• Repository: github.com/benoitpetit/prompt-my-project
• Issues: Report bugs or request features
• Releases: Download latest version
• Donate: Support on Liberapay

---