# Second Opinion

> Use when you want an independent review of an architecture decision, implementation choice, or technical approach before committing. Triggers on: "second opinion", "review this approach", "is this the right pattern", "validate my design", "Flutter/Riverpod architecture review".

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

---


# Second Opinion

When invoked:

1. **Summarize the problem** from conversation context (~100 words)

2. **Spawn subagent** using Task tool:
   - `gemini-consultant` with the problem summary

3. **Apply Flutter/Riverpod-specific checks**:
   - `ref.watch()` only in `build()`, `ref.read()` only in callbacks
   - `FooNotifier` generates `fooProvider` (not `fooNotifierProvider`)
   - No business logic in widgets — use notifiers/providers
   - `.select()` when only one field is consumed
   - GoRouter: no imperative `Navigator.push` inside providers

4. **Present combined results** showing:
   - Gemini's perspective
   - Flutter/Riverpod-specific findings
   - Where they agree/differ
   - Recommended approach with rationale

## Prerequisites

**Requires Gemini CLI installed locally.** This skill intentionally retains a Gemini CLI runtime dependency — it is the only Gemini reference remaining in this repo after the 2.0.0 Claude-only refactor.

Install: https://github.com/google/gemini-cli

## CLI Commands Used by Subagents

```bash
gemini -p "I'm working on a coding problem... [problem]"
```