A pip-installable CLI tool for API monitoring, analytics, and route improvement suggestions.

• Proxy-based Monitoring: Sits between clients and your API server to capture all traffic
• Hit Count Tracking: See which routes are most frequently accessed
• Request/Response Capture: Full details of every request and response
• Analytics Dashboard: Rich terminal UI showing performance metrics
• Advanced Pattern Detection: Automatically suggests improvements for 'chatty' APIs, slow writes, and security anomalies
• ASCII Graphs: Visualize request activity over time via CLI
• Dual UI: React+Ink UI for live monitoring, Textual fallback for pure Python
• Automated Testing: Comprehensive test suite for storage and analytics

pip install apimon

1. Start the proxy server (forwarding to your API on port 3000):

apimon proxy --target-port 3000

2. In another terminal, make some API requests to http://localhost:8080 (the proxy):

curl http://localhost:8080/api/users

3. View the dashboard:

apimon dashboard

apimon proxy --target-host localhost --target-port 3000 --port 8080

• --target-host: Your API server host (default: localhost)
• --target-port: Your API server port (default: 3000)
• --port: Proxy listen port (default: 8080)
• --db-path: Path to SQLite database (default: apimon.db)

apimon dashboard
apimon stats --hours 24
apimon requests --limit 50
apimon suggestions

• Python Backend: Proxy server using aiohttp, data storage with SQLite/SQLAlchemy
• Node.js Frontend: React+Ink CLI UI for interactive dashboards
• Textual Fallback: Pure Python TUI when Node.js is unavailable

All data is stored in a local SQLite database (apimon.db by default). The schema includes:

• requests: Individual request/response records
• route_stats: Aggregated statistics per route

The proxy automatically normalizes routes like:

• /users/123 → /users/{id}
• /posts/abc123 → /posts/{id}

This allows aggregating statistics for parameterized routes.

• APIMON_PORT: Port for the proxy server (default: 8080)
• APIMON_DB_PATH: Path to database file
• APIMON_LISTEN_HOST: Proxy listen host (default: 127.0.0.1, set to 0.0.0.0 for Docker)

You'll need 4 terminals:

Terminal 1 — start your API server (or use the built-in dummy API for testing):

uv run python scripts/dummy_api.py

Terminal 2 — build and start the apimon proxy:

docker compose up --build

The proxy starts on http://localhost:8080 and forwards to your API on port 3000. On Mac/Windows host.docker.internal resolves to your host machine automatically. On Linux, extra_hosts: host-gateway in docker-compose.yml handles it.

Terminal 3 — send requests through the proxy (or use the simulator):

uv run --with requests python scripts/simulate_requests.py

Terminal 4 — view analytics after traffic accumulates:

docker compose run --rm apimon apimon --db-path /data/apimon.db dashboard
docker compose run --rm apimon apimon --db-path /data/apimon.db stats
docker compose run --rm apimon apimon --db-path /data/apimon.db suggestions

Edit the command in docker-compose.yml:

command: apimon --db-path /data/apimon.db proxy --target-host my-api.internal --target-port 8000

docker build -f cicd/Dockerfile -t apimon .

docker run -d \
  -p 8080:8080 \
  -v apimon-data:/data \
  -e APIMON_LISTEN_HOST=0.0.0.0 \
  --add-host host.docker.internal:host-gateway \
  apimon apimon --db-path /data/apimon.db proxy --target-host host.docker.internal --target-port 3000

The SQLite database is stored in a Docker volume (apimon-data) mounted at /data inside the container. All analytics data survives container restarts. To wipe it:

docker compose down -v

# Install in development mode
pip install -e .

# Install frontend dependencies
cd frontend && npm install

# Run the proxy
apimon proxy --target-port 3000

# Run the UI
apimon ui

# Or use the Textual UI directly
apimon dashboard

MIT