CtrlK
BlogDocsLog inGet started
Tessl Logo

uinaf/cross-harness-review

Run a read-only code review through the opposite AI coding harness and return uinaf review-style findings, evidence, unverified gaps, and a ship-it / needs-review / blocked verdict. Use for /review-with-claudex, cross-harness review, opposite-model review, Claude reviewing Codex work, or Codex reviewing Claude work.

97

Quality

97%

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

headless-modes.mdreferences/

Headless Modes

Use this reference only when command choice or flags matter.

Claude Code

  • Headless entrypoint: claude -p or claude --print
  • Structured output: add --output-format json or --output-format stream-json
  • Fresh throwaway run: add --no-session-persistence
  • Avoid --bare by default; it skips CLAUDE.md auto-discovery and other context loading
  • Avoid claude ultrareview by default; it is hosted, can be expensive, and may not carry the local uinaf review contract unless explicitly prompted

Codex

  • General headless entrypoint: codex exec
  • Review headless entrypoint: codex exec review
  • Simple review alias: codex review
  • Throwaway run: add --ephemeral
  • Event stream: add --json
  • Last response capture: add -o <path> or --output-last-message <path>
  • Use codex exec review - when you need to inject the uinaf review contract or custom focus text

OpenAI Claude Code Plugin

OpenAI's Claude Code plugin can provide /codex:review. It is fine for a plain native Codex review from Claude Code, but prefer codex exec review - when this skill needs custom prompt text or local output handling.

references

headless-modes.md

SKILL.md

tile.json