paseo-advisor

Launches a single Paseo agent to give a second opinion on the current task.

15.7k|1.7k|Updated Oct 13, 2025
One-click install
npx skills add https://github.com/getpaseo/paseo --skill paseo-advisor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: paseo-advisor
Source: https://github.com/getpaseo/paseo/tree/main/skills/paseo-advisor
Command: npx skills add https://github.com/getpaseo/paseo --skill paseo-advisor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

When you are deep in a task, it is hard to get an unbiased outside perspective without handing off the work entirely. This Skill spins up a separate agent that reads your situation, gives a judgment, and leaves the decision-making to you.

Core Features & Use Cases

  • Profile-based advisor selection: Reads configured Paseo profiles via list_profiles and picks the one whose notes best match the question, optionally preferring a different provider family for a genuinely fresh take.
  • Self-contained briefing: Builds a zero-context prompt with the question, ruled-out options, relevant file paths, and an explicit no-edits constraint so the advisor only analyzes.
  • Forwarded skill execution: Passes through skill references like /unslop or /diagnose so the advisor runs that skill against the current task.
  • Persistent advisor mode: Keeps the advisor agent alive across multiple decisions when the user wants ongoing input.
  • Use Case: You are unsure whether to refactor a module or patch it. Invoke the advisor with the tradeoffs; it reads the relevant files and returns a reasoned recommendation without touching your code.

Quick Start

Ask the advisor for a second opinion on whether the current approach to the authentication refactor is sound.

Frequently Asked Questions about paseo-advisor

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

FAQPage Schema
How do I get a second opinion from another AI agent on my code?

Invoke the paseo-advisor skill with your question or topic. It selects a configured Paseo profile, briefs a separate agent with your context and relevant file paths, and returns the advisor's reasoned recommendation without letting it edit any files.

How does the advisor skill choose which agent profile to use?

It reads configured profiles via list_profiles and matches the profile notes to the question type, such as design, audit, or research. When several profiles fit, it prefers a different provider family so the second opinion is genuinely independent.

Can the advisor run another skill like a code review?

Yes. If the arguments contain a skill reference such as /unslop or /diagnose, the advisor is instructed to load and run that skill against the current task, with any remaining arguments passed through as the skill's input.

Does the advisor agent modify my files?

No. Every briefing ends with an explicit instruction that the work is analysis only and the advisor must not edit, create, or delete any files or write code. It only reads and returns a judgment.

Can I keep the same advisor for multiple decisions?

Yes. If you ask for ongoing input, the advisor agent is not archived after its first reply. You can send follow-up questions, and it is archived only when you say you are done or the topic shifts.