skills.rest
build000r
build000rCOMMUNITY
Explore all skills by this author →
Website𝕏@buildooorGitHub

prompt-reviewer

Score conversation prompts from Claude Code, Codex, AMP, and OpenCode using a 9-axis rubric.

5|Updated Feb 1, 2026
Search Tags:#codex#history#claude-code#trend#ai-prompting#prompt-quality#prompt-review
One-click install
npx skills add https://github.com/build000r/skills --skill prompt-reviewer
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: prompt-reviewer
Source: https://github.com/build000r/skills/tree/main/prompt-reviewer
Command: npx skills add https://github.com/build000r/skills --skill prompt-reviewer

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

The Prompt Review Skill enables you to quantify and improve AI prompting quality by automatically analyzing conversation prompts from Claude Code, Codex, and OpenCode, scoring them against a standardized rubric, and providing actionable feedback to drive better results over time.

Core Features & Use Cases

  • Comprehensive scoring: Applies a 9-axis rubric across sessions to yield a single composite score, with detailed axis insights.
  • Trend tracking: Maintains historical prompts reviews to visualize progress week over week and identify patterns.
  • Backfill & history: Supports backfilling past weeks and multiple providers to build a complete prompt-quality history.
  • Use cases: Teams or individuals aiming to benchmark prompting quality, refine conversation flows, and track improvements across Claude Code, Codex, AMP, and OpenCode.

Quick Start

Ask the agent to review your prompts and produce a composite score with trend history.

Frequently Asked Questions about prompt-reviewer

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

FAQPage Schema
How do I measure and score the quality of my AI prompts across conversations?▼

The system applies a 9-axis rubric to score your AI prompts, analyzing context and outcomes to yield a single composite score with detailed axis insights for tracking prompt quality improvements.

Can I review prompt quality for Claude Code and Codex conversation histories?▼

Yes, you can review prompt quality for Claude Code, Codex, AMP, and OpenCode. The system loads sessions directly from local ~/.claude and ~/.codex histories to analyze and score your interactions.

How do I track prompt quality trends week over week?▼

You can track prompt quality trends by saving review results to a history file. This maintains historical prompt reviews to visualize progress week over week and identify patterns over time.

Is it possible to backfill past weeks of prompt data for a complete quality history?▼

Yes, the system supports backfilling past weeks and multiple providers to build a complete prompt-quality history. This allows you to establish a baseline and populate missing historical data.

What is the best way to benchmark prompting quality for a team using multiple AI providers?▼

The best way to benchmark prompting quality is applying a standardized 9-axis rubric across multiple providers like Claude Code and Codex. Storing results in a history file enables teams to refine conversation flows and track improvements.

Related Skills

Explore More Content & Communication Skills→
moshehbenavrahammoshehbenavraham
2

session-logs

Search and analyze JSONL conversation histories to extract user and assistant messages.

Community
Intermediate
SoSoBurnerSoSoBurner

gstack-openclaw-retro

Analyze git commit history and code quality metrics for engineering retrospectives.

Community
Advanced
DicklesworthstoneDicklesworthstone
1.1k

coding_agent_session_search

Indexes and searches multi-agent coding session histories via TUI or CLI.

Community
Advanced
Ankh-StudioAnkh-Studio

prompt

Evaluate natural language prompts against a structured rubric with weighted scores.

Official
Intermediate