skills.rest
Ayoub-ouederni
Ayoub-ouederniCOMMUNITY
Explore all skills by this author →
GitHub

second-opinion

Runs external LLM code reviews on git diffs using Codex or Gemini CLI.

Updated Apr 3, 2026
Search Tags:#code review#security scan#codex cli#gemini cli#git diff#second opinion#llm review
One-click install
npx skills add https://github.com/Ayoub-ouederni/SENTINEL --skill second-opinion-ayoub-ouederni
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: second-opinion
Source: https://github.com/Ayoub-ouederni/SENTINEL/tree/main/.claude/skills/second-opinion
Command: npx skills add https://github.com/Ayoub-ouederni/SENTINEL --skill second-opinion-ayoub-ouederni

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Getting an independent code review from a different AI model helps catch issues that a single model might miss, but manually invoking external CLI tools with the right flags, prompts, and diff scopes is tedious and error-prone. ## Core Features & Use Cases - Dual-Model Reviews: Run OpenAI Codex CLI and Google Gemini CLI in parallel on the same diff, then compare where the two reviews agree and differ. - Flexible Review Scopes: Review uncommitted changes, branch diffs against the default branch, or a specific commit, with automatic default-branch detection and diff statistics preview. - Focused Reviews: Apply security, performance, or error-handling focus prompts, optionally include project conventions from CLAUDE.md or AGENTS.md, and run dependency scans only when manifest files changed. - Use Case: Before opening a pull request, ask for a second opinion on your branch diff with a security focus and receive structured findings from both Codex and Gemini with file and line references. ## Quick Start Ask for a second opinion review of your uncommitted changes using both Codex and Gemini with a general focus.

Frequently Asked Questions about second-opinion

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get a second opinion code review from another AI model?▼

Invoke the skill and choose a review tool, scope, and focus. It shells out to Codex CLI or Gemini CLI in headless mode, pipes the git diff with a review prompt, and returns structured findings with file and line references.

Codex CLI vs Gemini CLI for automated code review?▼

Codex exec produces structured JSON findings with priorities and confidence scores using an output schema, while Gemini CLI uses code-review and security extensions with freeform output. Running both in parallel lets you compare where the models agree and differ.

Can I review a branch diff instead of uncommitted changes?▼

Yes, the skill supports three scopes: uncommitted changes including untracked files, branch diffs against the auto-detected default branch, and a specific commit via git diff sha~1..sha. It shows diff statistics before running the review.

What happens if Codex or Gemini CLI is not installed?▼

The skill runs the selected tool directly and reports the install command if it fails, such as npm i -g @openai/codex or npm i -g @google/gemini-cli. If both were selected, it runs only the available tool and notes the skip.

When is the dependency security scan skipped?▼

The dependency scan runs only when the diff touches manifest files like package.json, requirements.txt, Cargo.toml, or go.mod. Since the scan analyzes the entire dependency tree regardless of scope, it is skipped when no dependency files changed.

Related Skills

Explore More Software Engineering Skills→
manuelbrandner85manuelbrandner85

second-opinion

Review git diffs with external LLM CLI tools like Codex and Gemini.

Community
Advanced
jovermierjovermier
1

quality-severity

Classify issues and code review findings into P1, P2, or P3 severity levels.

Community
Basic
uzumaki-incuzumaki-inc

health

Run code health checks and compute a weighted 0-10 score.

Official
Advanced
zli5460zli5460
2

blackbox

Automates software development workflows using a multi-model AI CLI.

Community
Intermediate