# Review

> Request a read-only adversarial Codex review with explicit model, effort, budget, and compact findings.

- Skill: `borda/review-4` (Agent Skill)
- Install (CLI): `npx skillmds@latest add borda/review-4`
- Raw SKILL.md: https://api.skillmd.com/api/skills/borda/review-4/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: Borda (https://skillmd.com/u/borda)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/borda/review-4

---


# Ask Codex to Review

Parse `$ARGUMENTS`: optional review instructions; optional `model`, `effort`, `timeout-seconds`, `depth`, `run-id`, `workspace`. Effort omitted: classify complete scope, pass selected level explicitly. Preserve caller-supplied level. Tiers: `low` = narrow mechanical change or settled fact; `medium` = bounded implementation, diagnosis, or review; `high` = cross-file, adversarial, architectural, or security judgment; `xhigh` = unusually broad and consequential; `max` = explicit caller request only.

Run `python "${CLAUDE_PLUGIN_ROOT}/bin/bridge_call.py" review --task "<instructions>"`; pass each supplied option separately. For quoted text you did not author, use scratch file + `--task-file <path>` instead of `--task`; mutually exclusive. Bridge runs read-only ephemeral general Codex execution with explicit adversarial-review prompt; default soft budget 300 seconds. Never resume review.

Return compact JSON envelope. Keep raw transcript at bridge-reported workspace-relative artifact path. Do not inline verbose peer `details`.