skills.rest
nklisch
nklischCOMMUNITY
Explore all skills by this author →
GitHub

tool-evaluator

Analyze AI-agent tool usage and generate a markdown report with severity ratings.

7|4|Updated Mar 2, 2026
Search Tags:#agent#evaluation#diagnostics#report#self-assessment#tool
One-click install
npx skills add https://github.com/nklisch/skills --skill tool-evaluator-nklisch
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: tool-evaluator
Source: https://github.com/nklisch/skills/tree/main/plugins/workflow/skills/tool-evaluator
Command: npx skills add https://github.com/nklisch/skills --skill tool-evaluator-nklisch

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It provides a systematic way for an AI agent to introspect its own tool usage, uncover confusion, inefficiencies, missing capabilities, and friction, and produce a concise, actionable report for tool authors and users.

Core Features & Use Cases

  • Structured self‑evaluation across six dimensions such as confusion points, efficiency, and API friction.
  • User interview to capture external feedback via AskUserQuestion.
  • Prioritized report generation with severity ratings, actionable recommendations, and a wishlist for new tools.
  • Use case: After a development session where the agent used Read, Write, or other tools, invoke this skill to assess and improve future tool interactions.

Quick Start

Ask the agent to run the tool evaluator for the current session.

Frequently Asked Questions about tool-evaluator

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

FAQPage Schema
How do I evaluate AI agent tool usage and find inefficiencies?▼

Self-assessment of agent tool usage analyzes interactions across a session to pinpoint inefficiencies, confusion points, and missing capabilities, producing a structured markdown report with severity ratings and actionable recommendations.

What is AI agent self-assessment for tool interactions?▼

Agent self-assessment systematically introspects tool usage across a conversation to uncover API friction, inefficiencies, and missing capabilities, yielding a concise actionable report for tool authors and users.

How do I generate a diagnostics report for agent tool friction?▼

You generate a diagnostics report by analyzing tool usage across six dimensions such as confusion points, efficiency, and API friction, which outputs a prioritized markdown summary with severity ratings and a wishlist for new tools.

Can I capture user feedback during an agent tool evaluation?▼

Yes, you can capture user feedback during an evaluation by conducting a user interview via AskUserQuestion, integrating external feedback into the final structured report alongside efficiency findings and actionable recommendations.

When should I run a diagnostics evaluation on my agent session?▼

You should run a diagnostics evaluation after a development session where the agent invoked tools like Read, Write, or AskUserQuestion, enabling self-assessment to assess and improve future tool interactions.

Related Skills

Explore More Software Engineering Skills→
ericdahl-devericdahl-dev

retro

Analyze commit history, work patterns, and code quality metrics for weekly engineering retrospectives.

Official
Intermediate
makyua-sanmakyua-san

devex-review

Audit developer experience by navigating docs, running getting started flows, and evaluating CLI help text.

Community
Advanced
Liberty91LTDLiberty91LTD
15

maturity-assessment

Assess CTI programme maturity across six dimensions with a structured scoring framework.

Official
Intermediate
zhaojun1zhaojun1

munger-lens

Analyze business opportunities using Munger's mental models and bias checklists.

Community
Advanced