skills.rest
usadamasa
usadamasaCOMMUNITY
Explore all skills by this author →
Website𝕏@usadamasaGitHub

prompt-review

Analyze AI agent conversation logs and generate a Markdown report in Japanese.

Updated Aug 27, 2026
Search Tags:#code review#prompt engineering#security audit#ai analysis#conversation log#technical assessment
One-click install
npx skills add https://github.com/usadamasa/claude-config --skill prompt-review-usadamasa
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prompt-review
Source: https://github.com/usadamasa/claude-config/tree/main/.agents/skills/prompt-review
Command: npx skills add https://github.com/usadamasa/claude-config --skill prompt-review-usadamasa

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires python3, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill analyzes your past AI chat interactions to help you understand your technical understanding, prompting patterns, and AI dependency.

Core Features & Use Cases

  • Analyze conversation logs: Reviews logs from various AI agents (Claude Code, Copilot Chat, etc.).
  • Estimate technical understanding: Assesses your grasp of concepts based on prompts.
  • Identify prompting patterns: Detects effective and improvable ways you interact with AI.
  • Gauge AI dependency: Evaluates how much you rely on AI for decision-making.
  • Generate detailed reports: Creates a Markdown report summarizing findings with evidence.
  • Use Case: After a month of using AI coding assistants, run this Skill to get a report on how your prompting has evolved and where you might be over-relying on AI.

Quick Start

Run the prompt-review skill to analyze your AI chat history for the last 30 days.

Frequently Asked Questions about prompt-review

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

FAQPage Schema
How do I analyze AI chat history to improve my prompt engineering?▼

Analyzing AI chat history involves reviewing conversation logs from AI agents to estimate technical understanding, identify prompting patterns, and assess AI dependency, resulting in a detailed Markdown report with evidence-based insights.

Can I review conversation logs from GitHub Copilot Chat and Claude Code together?▼

Reviewing conversation logs from GitHub Copilot Chat and Claude Code is supported. The Skill aggregates logs from these agents, plus Cline, Roo Code, Windsurf, Antigravity, and OpenCode, to provide a comprehensive technical assessment.

Do I need python3 installed to run an AI conversation log security audit?▼

Python3 is required to perform an AI conversation log security audit. The Skill uses python3 scripts to process chat history and detect secret credentials leaked during AI agent interactions.

What is the best way to assess AI dependency in coding assistants?▼

Assessing AI dependency in coding assistants is done by analyzing conversation history to evaluate how much you rely on AI for decision-making. This Skill gauges over-reliance by reviewing prompting patterns and technical understanding.

How do I filter AI agent conversation logs by project name and days?▼

Filtering AI agent conversation logs by project name and days narrows the technical assessment to a specific timeframe and codebase. This generates a focused Markdown report containing project-specific summaries and security warnings.

Related Skills

Explore More Software Engineering Skills→
bahayonghangbahayonghang
16

memory-system

Index Markdown notes into a local SQLite vector/full-text index.

Community
Intermediate
claude-worldclaude-world
81

project-health-check

Evaluate software project health across documentation, testing, security, code quality, dependencies, database, and deployment.

Official
Intermediate
robpitcherrobpitcher

secret-handling

Block agents from reading live .env credentials and writing secrets to committed files.

Community
Advanced
tokorotentokoroten
359

prompt-review

Analyze AI agent chat histories to estimate technical understanding and generate Japanese reports.

Community
Advanced