Skip to content

Add prepare eval gate#8

Merged
ryanbaumann merged 1 commit into
mainfrom
agent/agent-loop-ux
Jul 8, 2026
Merged

Add prepare eval gate#8
ryanbaumann merged 1 commit into
mainfrom
agent/agent-loop-ux

Conversation

@ryanbaumann

Copy link
Copy Markdown
Owner

What changed

  • Added a deterministic Prepare eval gate in the web app before image rendering.
  • Surfaced Prepare eval pass/warn status in the Studio thought stream.
  • Moved dynamic filename generation off the critical render path with a local slug fallback.
  • Synced the portable skill and npm CLI to the same PrepareResult contract and eval gate; bumped skill package metadata to v3.1.0.
  • Documented the audit findings and follow-up eval work in docs/agent-loop-audit.md.

Why

The agent loop already had a strong two-agent shape, but the handoff from Prepare to Render was only prompt-instructed, not locally enforced. This PR makes the handoff deterministic, keeps optional sidecars from failing successful renders, and keeps the portable skill/CLI aligned with the web app logic.

Validation

  • npm run lint && npm test && npm run build
  • npm run build
  • python3 -m py_compile skill/infographic-agent/portable_infographic.py
  • python3 skill/infographic-agent/portable_infographic.py --help
  • node --check skill/infographic-agent/bin/infographic-agent.js
  • node skill/infographic-agent/bin/infographic-agent.js --help
  • direct/no-research Prepare eval smoke via Python import
  • skill-creator quick_validate.py skill/infographic-agent with PyYAML installed in a temporary PYTHONPATH

@ryanbaumann ryanbaumann marked this pull request as ready for review July 8, 2026 23:22
@chatgpt-codex-connector

Copy link
Copy Markdown

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.

@ryanbaumann ryanbaumann merged commit 34a799f into main Jul 8, 2026
2 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant