SE

second-opinion-start

Reference card for ponytail commands. Use this to check modes and skills for minimalist development.

Install

mkdir -p .claude/skills/second-opinion-start && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/18847" && unzip -o skill.zip -d .claude/skills/second-opinion-start && rm skill.zip

Installs to .claude/skills/second-opinion-start

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Obtain a host-managed multi-model review of a file, diff, or design question
76 charsno explicit “when” trigger
Intermediate

Key capabilities

  • Confirm `codex mcp get second-opinion` success
  • Read explicitly requested file or diff for review
  • Use `get_code_second_opinion` for default model review
  • Use `get_multi_model_second_opinion` for multiple model review
  • Return a concise synthesis of agreements, risks, and recommendations

How it works

This skill obtains a multi-model review of code or design questions by using a service's second opinion tools and synthesizing the results.

Inputs & outputs

You give it
A file, diff, or design question for review, along with language and requested depth
You get back
A concise synthesis of agreements, disagreements, risks, recommended action, model names, and reported cost

When to use second-opinion-start

  • Get help on ponytail commands
  • Check current mode
  • Audit code debt
  • Display mode reference

About this skill

Start a Second Opinion

  1. Confirm codex mcp get second-opinion succeeds. If it does not, use the exact graceful-degradation message from $second-opinion-help and continue with a normal Codex review only if the user still wants one.
  2. Read the explicitly requested file or diff. Never include .env, key files, credentials, or unredacted logs in the consultation payload.
  3. Use the service's get_code_second_opinion tool for its default model or get_multi_model_second_opinion when the user selects one or more models. Supply code/diff, language, review question, and requested depth.
  4. Return a concise synthesis: agreements, disagreements, risks, recommended action, model names, and reported cost. Do not treat a model response as an authority over repository evidence.

For a real code-review dogfood run, inspect a non-secret current diff and ask for focused review of the changed behavior. Record only the non-sensitive synthesis in the issue or PR.

When not to use it

  • When the user asks to include `.env` files in the consultation payload
  • When the user asks to include key files or credentials
  • When the user asks to include unredacted logs

Limitations

  • Does not include `.env`, key files, credentials, or unredacted logs in the consultation payload
  • Does not treat a model response as an authority over repository evidence
  • Requires `codex mcp get second-opinion` to succeed

How it compares

This skill provides a synthesized multi-model perspective on code or design, offering a broader range of feedback and risk identification compared to a single model review.

Compared to similar skills

second-opinion-start side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
second-opinion-start (this skill)017dReviewIntermediate
webapp-testing3533moReviewIntermediate
resolve-conflicts818moReviewIntermediate
telegram-bot-builder1066moReviewIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

webapp-testing

anthropics

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

353585

resolve-conflicts

antinomyhq

Use this skill immediately when the user mentions merge conflicts that need to be resolved. Do not attempt to resolve conflicts directly - invoke this skill first. This skill specializes in providing a structured framework for merging imports, tests, lock files (regeneration), configuration files, and handling deleted-but-modified files with backup and analysis.

81334

telegram-bot-builder

davila7

Expert in building Telegram bots that solve real problems - from simple automation to complex AI-powered bots. Covers bot architecture, the Telegram Bot API, user experience, monetization strategies, and scaling bots to thousands of users. Use when: telegram bot, bot api, telegram automation, chat bot telegram, tg bot.

106130

dev-browser

SawyerHood

Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include "go to [url]", "click on", "fill out the form", "take a screenshot", "scrape", "automate", "test the website", "log into", or any browser interaction request.

53176

openspec-onboard

studyzy

Guided onboarding for OpenSpec - walk through a complete workflow cycle with narration and real codebase work.

10207

codex-cli-bridge

alirezarezvani

Bridge between Claude Code and OpenAI Codex CLI - generates AGENTS.md from CLAUDE.md, provides Codex CLI execution helpers, and enables seamless interoperability between both tools

9180

Search skills

Search the agent skills registry