🔧 Fixit.AI

AI-Powered API Testing Agent with Zero-Config Onboarding for OpenAPI-First Development

Fixit.AI is an offline-first API testing platform that automatically generates comprehensive test suites, detects security vulnerabilities, and provides AI-powered failure analysis. Built for modern API development with native support for FastAPI and Express.js.

🌟 Key Features

• 🚀 Zero-Config Onboarding - Start testing in seconds with automatic spec discovery
• 🤖 AI-Powered Analysis - Intelligent failure detection and code suggestions
• 🔒 Security-First - Built-in vulnerability scanning and OWASP compliance
• 📊 Rich Reporting - Beautiful HTML reports with detailed analytics
• ⚡ Offline-First - Works entirely on your machine with local LLMs
• 🛠️ Smart Patching - Automated code fixes for common API issues

🎯 Perfect For

API Developers • QA Engineers • DevOps Teams • Startups

---

⚡ Quick Start

Installation

pip install fixit-ai

or 

# Clean installation (recommended)
pip install fixit-ai --quiet

FastAPI (Zero-Config)

# Start your FastAPI app
uvicorn main:app --reload

# Test with zero configuration
fixit init --fastapi main:app --base http://localhost:8000
fixit gen && fixit test && fixit fix && fixit sec

Express.js

# Start your Express.js app with OpenAPI
node app.js

# Test your Express.js API
fixit init --express . --base http://localhost:3000
fixit gen && fixit test && fixit fix && fixit sec

Universal (Any OpenAPI Framework)

fixit init --spec openapi.yaml --base http://localhost:8000
fixit gen && fixit test && fixit fix && fixit sec

🎯 What Fixit.AI Does

Step	Command	What Happens
Initialize	fixit init	🔍 Auto-extracts OpenAPI spec, creates fixit.toml with LLM defaults
Generate	fixit gen	📝 Creates comprehensive test suite covering all endpoints
Test	fixit test	🧪 Runs tests with real-time AI analysis of failures
Fix	fixit fix	🛠️ Shows AI-generated code patches (suggestions-only by default)
Secure	fixit sec	🔒 Scans for OWASP API Security Top 10 vulnerabilities

🤖 AI Integration

Fixit.AI works with local LLMs for private, offline analysis:

Quick Setup (LM Studio)

1. Download LM Studio
2. Install model: GPT-OSS 20B
3. Start server on localhost:1234
4. Run: fixit test (auto-detects LM Studio)

Alternative (Ollama)

# Install and setup
curl -fsSL https://ollama.ai/install.sh | sh
ollama pull phi3:mini

# Configure (optional - auto-detected)
fixit init --llm-provider ollama --llm-model phi3:mini

📋 Framework Support

Framework	Language	Zero-Config	Code Patches	Security Scan
FastAPI	Python	✅	✅	✅
Express.js	Node.js	✅	✅	✅
Any OpenAPI 3.0+	Any	✅	✅	✅

⚙️ Configuration

Auto-Generated fixit.toml (created on first run):

llm_active = "llm"
openapi = "openapi.yaml"

[llm]
provider = "lmstudio"
# base_url = "http://localhost:11434/v1" # Ollama
base_url = "http://localhost:1234/v1"    # LM Studio
model = "gpt-oss20B"
temperature = 0.0
max_tokens = 2048
timeout_seconds = 180
offline = true
api_key_env = "FIXIT_API_KEY"

[llm.demo]
provider = "llama.cpp-stub"

[test]
include = []
exclude = []
max_tests = 0
concurrency = 8

Override during init:

fixit init --fastapi main:app --base http://localhost:8000 \
  --llm-provider ollama --llm-model phi3:mini

🛠️ CLI Commands

Command	Description
fixit init --fastapi main:app	Initialize FastAPI project (zero-config)
fixit init --express .	Initialize Express.js project
fixit init --spec openapi.yaml	Initialize with existing OpenAPI spec
fixit gen	Generate comprehensive test suite
fixit test	Run tests with AI analysis
fixit fix	Show AI code suggestions
fixit sec	Security vulnerability scan

🚀 Getting Started Checklist

• Install: pip install fixit-ai
• Choose framework: FastAPI (zero-config) or Express.js
• Initialize: fixit init --fastapi main:app --base http://localhost:8000
• Generate tests: fixit gen
• Run tests: fixit test
• Review suggestions: fixit fix
• Security scan: fixit sec
• View report: Open .fixit/reports/index.html

🔍 Example Output

Real-time AI Analysis:

$ fixit test
🧪 Running 7 tests...
❌ 3 failures detected

🤖 AI Analysis:
[1/3] Analyzing post_/users_validation_error... 🆕 (fresh)
[2/3] Analyzing get_/users/{id}_not_found... 💾 (cached)
[3/3] Analyzing post_/auth/login_unauthorized... 🆕 (fresh)

$ fixit fix
💡 3 AI-generated suggestions:
✨ Fix email validation in UserCreate model
✨ Add proper 404 error handling
✨ Implement JWT token validation

🔗 Resources

• 📚 Full Documentation - Comprehensive guides and tutorials
• 🐛 Issues - Bug reports and feature requests
• 💬 Discussions - Community support and Q&A
• ⭐ GitHub - Source code and examples

📄 License

MIT License - see LICENSE for details.

---