Skip to content

fix: preserve runtime credentials in background review fork#2581

Closed
mwotton wants to merge 1 commit into
NousResearch:mainfrom
mwotton:fix/background-review-runtime
Closed

fix: preserve runtime credentials in background review fork#2581
mwotton wants to merge 1 commit into
NousResearch:mainfrom
mwotton:fix/background-review-runtime

Conversation

@mwotton

@mwotton mwotton commented Mar 23, 2026

Copy link
Copy Markdown

Summary

  • pass the active runtime credentials into the background memory/skill review agent
  • keep the review fork on the same endpoint and auth as the main agent
  • add a regression test covering custom OpenAI-compatible endpoints

Root cause

The background review fork only inherited model and provider. For custom OpenAI-compatible routes like Local Qwen, Hermes stores the transport provider as openrouter, so the fork would fall back to https://openrouter.ai/api/v1 with no API key.

Testing

  • PYTEST_DISABLE_PLUGIN_AUTOLOAD=1 PYTHONPATH=/home/mark/.hermes/hermes-agent-bg-review /home/mark/.hermes/hermes-agent/venv/bin/pytest -c /dev/null /home/mark/.hermes/hermes-agent-bg-review/tests/test_background_review_runtime.py -q
  • python -m py_compile /home/mark/.hermes/hermes-agent-bg-review/run_agent.py

@teknium1

Copy link
Copy Markdown
Contributor

Thanks for this contribution, @mwotton! The bug you identified is real and well-described.

This is an automated hermes-sweeper review.

The fix has since been independently implemented and merged to main:

Closing as implemented on main.

@teknium1 teknium1 closed this Apr 27, 2026
@alt-glitch

Copy link
Copy Markdown
Collaborator

Superseded by #16099 (merged) which fixes the same root cause — background review fork inheriting parent's live runtime credentials.

@alt-glitch alt-glitch added type/bug Something isn't working P3 Low — cosmetic, nice to have comp/agent Core agent loop, run_agent.py, prompt builder duplicate This issue or pull request already exists labels May 3, 2026
@alt-glitch

Copy link
Copy Markdown
Collaborator

Superseded by #16099 (merged)

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

Labels

comp/agent Core agent loop, run_agent.py, prompt builder duplicate This issue or pull request already exists P3 Low — cosmetic, nice to have type/bug Something isn't working

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants