Verify Endpoints

Automated Endpoint Verification Playbook

This skill instructs the agent to run and test local development servers and verify the responsiveness, response format, and stability of endpoints after any change.

Target Triggers

• /verify-endpoints
• /loop-debug
• "verify endpoint works on dev server"

Action Plan

1. Identify Server Start Commands

• Find how the server is booted locally (e.g. npm run dev, python main.py, flask run, uvicorn main:app, go run main.go, rails s).
• Determine the port and host configuration (typically http://localhost:3000 or http://127.0.0.1:8000).

2. Launch Local Dev Server in Background

• Run the server boot command as a background task using run_command with an asynchronous wait time or separate terminal output.
• Wait a few seconds for the database/server connections to initialize.
• Monitor log files or stdout/stderr logs to ensure the server starts without compilation/boot errors.

3. Probe and Verify Endpoints

• Execute target curl/HTTP requests representing standard usage:
  • Check the health check endpoint (e.g. /health, /).
  • Perform GET, POST, or PUT actions with test payloads on modified routes.
  • Review the HTTP response headers and status codes (e.g. 200 OK, 201 Created).
  • Parse the JSON body to verify schema correctness.

4. Handle Failures & Loop Debug

• If a request returns 500 Internal Server Error, 404 Not Found, or fails to connect:
  • Inspect server stdout/stderr logs for traces, exceptions, database errors, or syntax issues.
  • Fix the code matching the traceback.
  • Re-run curl tests until everything passes cleanly.

5. Cleanup

• Terminate the background server task once testing is complete using manage_task with action kill.
• Clean up any temporary or mock database states created during manual probes.