Get Codex's code review of git changes after Claude makes edits. Trigger when user wants a second opinion on code changes ("have Codex review my changes", "get code review from Codex", "review this diff with Codex"), or as a final check before committing.
/plugin marketplace add robbyt/claude-skills/plugin install codex@robbyt-claude-skillsThis skill inherits all available tools. When active, it can use any tool Claude has access to.
Have Codex review git changes for a second perspective on code quality.
Every prompt sent to Codex MUST include these instructions:
"You are running non-interactively as part of a script. Do not ask questions or wait for input. Do not make any changes. Provide your complete feedback immediately."
Codex is a consultant. Claude Code handles all file modifications.
If the codex MCP tool is available, first save the diff then review:
git diff --cached > codex-review.diff
mcp__plugin_codex_cli__codex({
"prompt": "You are running non-interactively as part of a script. Do not ask questions or wait for input. Do not make any changes. Provide your complete feedback immediately.\n\nReview the code changes at codex-review.diff for bugs, security issues, and style problems.",
"sandbox": "read-only",
"model": "gpt-5.2"
})
rm codex-review.diff
If MCP is unavailable, use shell commands:
git diff --cached > codex-review.diff
codex exec "You are running non-interactively as part of a script. Do not ask questions or wait for input. Do not make any changes. Provide your complete feedback immediately.
Review the code changes at codex-review.diff for issues." --sandbox read-only -m gpt-5.2-codex 2>&1
rm codex-review.diff
Or use the built-in review command:
codex review --uncommitted 2>&1
Note: The review command is scoped to diffs and doesn't support --sandbox.
Staged changes:
mcp__plugin_codex_cli__codex({
"prompt": "You are running non-interactively as part of a script. Do not ask questions or wait for input. Do not make any changes. Provide your complete feedback immediately.\n\nReview codex-review.diff for:\n1. Bugs or logic errors\n2. Security vulnerabilities\n3. Style inconsistencies\n4. Missing error handling",
"sandbox": "read-only",
"model": "gpt-5.2"
})
Security focus:
mcp__plugin_codex_cli__codex({
"prompt": "You are running non-interactively as part of a script. Do not ask questions or wait for input. Do not make any changes. Provide your complete feedback immediately.\n\nSecurity review of codex-review.diff. Check for:\n- XSS vulnerabilities\n- SQL/command injection\n- Sensitive data exposure\n- Authentication/authorization issues",
"sandbox": "read-only",
"model": "gpt-5.2"
})
Performance focus:
mcp__plugin_codex_cli__codex({
"prompt": "You are running non-interactively as part of a script. Do not ask questions or wait for input. Do not make any changes. Provide your complete feedback immediately.\n\nPerformance review of codex-review.diff. Check for:\n- Inefficient algorithms\n- N+1 queries\n- Memory leaks\n- Blocking operations",
"sandbox": "read-only",
"model": "gpt-5.2"
})
sandbox: "read-only" to prevent file modificationssandbox: "danger-full-access" - this is forbiddendangerouslyDisableSandbox: truereferences/setup.md for troubleshooting