recommending-skills

Score installed skills against user requests using keyword and context matching.

1|Updated Feb 19, 2026
One-click install
npx skills add https://github.com/Open330/agt --skill recommending-skills
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: recommending-skills
Source: https://github.com/Open330/agt/tree/main/meta/skill-recommender
Command: npx skills add https://github.com/Open330/agt --skill recommending-skills

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Automatically identifies installed skills that could assist with the user's current request and presents concise recommendations without automatically executing them.

Core Features & Use Cases

  • Non-intrusive auto-suggestion: scans the skill inventory to surface relevant capabilities based on user intent.
  • Keyword and context matching: compares request text with each skill's description to rank candidates.
  • Real-world use: when you ask for a feature, the agent proposes skills like git-commit-pr or proposal-analyzer to consider next steps.

Quick Start

Ask the agent to surface recommended skills when making a request.

Frequently Asked Questions about recommending-skills

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

FAQPage Schema
How do I automatically recommend relevant skills based on user request context?

To recommend relevant skills, this tool scans the installed skill inventory and uses keyword and context-based matching with frontmatter metadata to surface top recommendations based on user intent.

Does the skill recommender automatically execute suggested skills?

No, the skill recommender does not automatically execute suggested skills; it only scores potential matches from the inventory scan and surfaces non-intrusive recommendations for the user to consider.

How does keyword matching handle noise when surfacing agent-skills?

Keyword matching handles noise by applying a configurable sensitivity setting during context analysis, balancing relevance and noise to accurately rank appropriate skills from the local inventory script.

What is the best way to surface installed skills for an automation workflow?

The best way to surface installed skills is to let the agent analyze your request text, compare it against frontmatter metadata, and present concise skill recommendations without triggering automatic execution.

Can I use context-analysis to find skills without installing additional dependencies?

Yes, you can use context-analysis to find skills without additional dependencies, as the matching operates locally using a built-in inventory script and existing frontmatter metadata.

Why are irrelevant skills appearing in my skill inventory scan results?

Irrelevant skills appear in inventory scan results when the matching sensitivity is configured too loosely; adjusting the configurable sensitivity setting will balance relevance and reduce noise.