Skip to content

Rate limit and context usage visibility #75

Open

Description

@gopaljigaur

Provider errors only reach on_llm_error and RunResult.error, and there is no way to see how full the context is.

Proposal An event carrying the provider's rate limit headers and the HTTP status on errors, and a method reporting a session's token count against the model window and the compaction threshold.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    enhancementNew feature or request

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions