CtrlK
BlogDocsLog inGet started
Tessl Logo

finsi/codex-review

Use when the user wants a local second-opinion code review via OpenAI Codex CLI — on the current branch, staged changes, a single file, or a piped diff. Triggers include "codex review", "review with codex", "run codex on this diff", "second opinion from codex", or pre-commit / pre-PR review requests that ask for codex specifically.

90

2.25x
Quality

97%

Does it follow best practices?

Impact

97%

2.25x

Average score across 2 eval scenarios

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

criteria.jsonevals/scenario-1/

{
  "context": "Tests whether the agent correctly invokes a staged-only pre-commit Codex review: using 'git diff --staged' to scope the diff, writing to a /tmp/ file first, piping via stdin (not inline), passing PROMPT.md, using --sandbox read-only, and capturing output.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Staged-only diff command",
      "description": "Uses 'git diff --staged' (not 'git diff HEAD' or 'git diff main...HEAD') to capture only staged changes",
      "max_score": 15
    },
    {
      "name": "Write diff to /tmp/",
      "description": "Writes the diff output to a file under /tmp/ (e.g., /tmp/codex-review.diff) before passing it to codex",
      "max_score": 15
    },
    {
      "name": "Pipe via stdin not inline",
      "description": "Passes the diff to codex exec via stdin pipe (cat /tmp/... | codex exec ...) rather than embedding the diff inline in the command string",
      "max_score": 15
    },
    {
      "name": "PROMPT.md passed",
      "description": "Passes the contents of .claude/skills/codex-review/resources/PROMPT.md as the prompt argument to codex exec",
      "max_score": 15
    },
    {
      "name": "--sandbox read-only flag",
      "description": "Includes --sandbox read-only in the codex exec invocation",
      "max_score": 10
    },
    {
      "name": "Output captured",
      "description": "Redirects codex output to a file (e.g., /tmp/codex-review.out) rather than only printing to terminal",
      "max_score": 15
    },
    {
      "name": "CRITICAL/HIGH surfacing",
      "description": "Script or accompanying comment indicates that CRITICAL and HIGH findings should be surfaced / reviewed after capture",
      "max_score": 15
    }
  ]
}

evals

scenario-1

criteria.json

task.md

SKILL.md

tile.json