Skip to content

ouroboros pm hardcodes Claude model instead of using configured clarification model #240

@srausser

Description

@srausser

Summary

ouroboros pm ignores the configured clarification model and starts with a hardcoded Claude model string.

Repro

  1. Install ouroboros-ai 0.26.3 and run ouroboros setup with Codex runtime.
  2. Confirm ~/.ouroboros/config.yaml has llm.backend: codex.
  3. Run ouroboros pm.

Actual

The command prints:

Model: anthropic/claude-sonnet-4-20250514

This happens even when the runtime/backend is configured for Codex and even when the clarification model in config should resolve to a backend-safe default.

Expected

ouroboros pm should source its default from config, ideally via get_clarification_model(), instead of hardcoding an Anthropic model in the CLI option default.

Cause

In ouroboros/cli/commands/pm.py, the callback currently defines:

] = "anthropic/claude-sonnet-4-20250514"

That bypasses the config loader entirely.

Local workaround

I patched my local install to import get_clarification_model and use that as the default for the --model option. After that, ouroboros pm starts with Model: default under Codex as expected.

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugReproducible defect or broken behavior

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions