Hey, thanks for the report and the request ID.
Your hypothesis is right. This isn’t a schema exposure bug the schema is passed to the model correctly, including required fields. It’s the model’s instruction-following ability. In auto mode, after exceeded usage, requests get routed to lighter models. Those models sometimes make up parameter names like query instead of task, even when the schema clearly says otherwise.
What you can do:
- Pick a specific model instead of
autolike Claude Sonnet or GPT-5. They follow MCP schemas a lot better. - In the prompt, explicitly name the fields like
use codegraph-context with task="...". This greatly reduces mistakes on weaker models.
Related threads: