Find out exactly why
your AI agent broke
Paste logs or upload config files. Get a plain-English diagnosis, root cause, and fix plan in seconds.
Debugging AI agents is painful
Cryptic logs, errors from a dozen failure points, and no tool built for this exact problem.
Logs are cryptic
Stack traces, error codes, and provider-specific jargon that means nothing out of context.
Errors come from many layers
Is it your config? The model provider? A gateway? Permissions? It could be any of a dozen failure points.
Chatbots guess, Medic diagnoses
Pasting logs into ChatGPT gives generic advice. Medic matches known failure patterns and cites evidence.
Three steps to a diagnosis
No sign-up. No terminal commands. No guesswork.
Paste your logs
Copy terminal output, error logs, or stack traces. Upload config files too.
Pattern matching
25+ diagnostic rules scan the parsed data and extract supporting evidence.
Confidence scored
Each diagnosis gets a confidence level based on how many signals matched.
Get the fix
One recommended fix, alternatives, and what to avoid — with official docs linked.
See a real diagnosis
This is what Medic returns when it detects an invalid API key.
Invalid or Rejected API Key
The provider rejected the API key with a 401 Unauthorized. The key may be expired, incorrect, or belong to a different environment.
✓ Regenerate the API key at platform.openai.com/api-keys
Works with your stack
Medic auto-detects your tools and providers from the logs.
Ready to fix your agent?
Paste your logs, get your diagnosis. It takes less than 10 seconds.
Analyze Logs