mentor

Identify missing AI workflows from repository and Claude Code configuration signals.

4|2|Updated Jun 26, 2026
One-click install
npx skills add https://github.com/gwenneg/ai-mentor --skill mentor-gwenneg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mentor
Source: https://github.com/gwenneg/ai-mentor/tree/main/skills/mentor
Command: npx skills add https://github.com/gwenneg/ai-mentor --skill mentor-gwenneg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps developers discover AI and Claude Code capabilities they are not yet using, without requiring them to know what to ask for. It analyzes the current conversation, repository, configuration signals, and a persistent local profile to provide a personalized, grounded recommendation while avoiding repeated or declined suggestions.

Core Features & Use Cases

  • Personalized capability discovery: Identifies the highest-leverage workflow missing from a developer's current practices.
  • Repo-grounded recommendations: Uses real project files, commands, configuration, and session signals to create actionable advice rather than generic guidance.
  • Persistent learning profile: Records capabilities as shown, adopted, or declined so recommendations improve over time and are not repeated.
  • Demonstrated next actions: Produces ready-to-use prompts and offers to configure supported workflows directly, including hooks, agents, skills, MCP integrations, CI automation, and review workflows.
  • Use Case: A developer asks how to improve testing or invokes the mentor without a question; the Skill inspects the repository and known Claude Code setup, selects one relevant capability, explains the principle behind it, and gives a concrete next step.

Quick Start

Invoke the mentor with a specific AI workflow question or leave the argument empty to discover the most valuable Claude Code capability you are not using yet.

Frequently Asked Questions about mentor

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

FAQPage Schema
How do I discover new AI workflows for Claude Code?

To discover new AI workflows for Claude Code, you can analyze your repository files and configuration signals to receive personalized recommendations for high-leverage capabilities you are not using yet. This avoids generic guidance by using real project context.

What is the best way to improve developer productivity with AI automation?

The best way to improve developer productivity with AI automation is by identifying the most valuable capability missing from your current practices. Analyzing session signals and a persistent learning profile helps generate ready-to-use prompts and configure workflows directly.

Can I get personalized AI debugging and testing recommendations based on my codebase?

Yes, you can get personalized AI debugging and testing recommendations by granting read-only access to your repository and Claude Code configuration. The system inspects real project files and session signals to provide actionable advice tailored to your setup.

Does Claude Code capability discovery remember previously declined AI workflow suggestions?

Yes, Claude Code capability discovery remembers previously declined or adopted suggestions. It uses a persistent local profile that records capabilities as shown, adopted, or declined, ensuring recommendations improve over time and are not repeated.

How do I configure CI automation and code review workflows using AI?

To configure CI automation and code review workflows using AI, you can use demonstrated next actions that produce ready-to-use prompts. These actions offer to set up supported workflows directly, including hooks, agents, skills, and MCP integrations.

What access is needed to suggest non-repeating AI development workflows?

Suggesting non-repeating AI development workflows requires read-only access to relevant repository and Claude Code configuration signals, plus edit access to a local profile. This combination enables personalized recommendations without repeating past suggestions.