meta-review

Audit diaries and reflections to identify recurring errors and automation opportunities.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/vitalwarley/hyprdots --skill meta-review-vitalwarley
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: meta-review
Source: https://github.com/vitalwarley/hyprdots/tree/main/claude-global/skills/meta-review
Command: npx skills add https://github.com/vitalwarley/hyprdots --skill meta-review-vitalwarley

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill audits session diaries and reflections to surface recurring errors, wasted effort, instruction drift relative to CLAUDE.md rules, and repeated manual steps that can be automated so you can prioritize improvements and reduce future friction.

Core Features & Use Cases

  • Error recurrence detection: groups and counts repeated failures across sessions, identifies root causes, and checks for existing rules that address them.
  • Instruction drift analysis: compares diary decisions and reflection violations against CLAUDE.md rules to flag conflicting, weakened, or dormant rules.
  • Automation candidate discovery: finds repeated manual steps suitable for hooks, skills, or rules and drafts proposals for automation.
  • Use Case: run a weekly or deep review to produce an evidence-backed meta-review that recommends CLAUDE.md edits and new automation hooks.

Quick Start

Run /meta-review last week to generate an evidence-backed meta-review of recent sessions and save it to your reviews folder.

Frequently Asked Questions about meta-review

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

FAQPage Schema
How do I audit my workflow diaries to find recurring errors?

To audit workflow diaries for recurring errors, you need a meta-review process that groups and counts repeated failures across sessions, identifies root causes, and checks existing rules. This Skill automates that by analyzing local memory files and producing an evidence-cited Markdown report.

What is instruction drift and how do I detect it in CLAUDE.md rules?

Instruction drift occurs when diary decisions and reflections conflict with CLAUDE.md rules. You detect it by comparing session notes against project rules to flag conflicting, weakened, or dormant rules, generating an evidence-backed meta-review of the discrepancies.

How can I identify automation candidates from my reflection notes?

You identify automation candidates by scanning reflection notes and diaries for repeated manual steps suitable for hooks, skills, or rules. This Skill discovers those repeated steps and drafts proposals for automation to reduce future friction.

Can I analyze local memory files to review personal project workflows?

Yes, you can analyze local memory files to review personal project workflows. This Skill applies to local ~/.claude memory files and project CLAUDE.md rules, scanning recent or deep time windows to surface wasted effort and automation opportunities.

What's the best way to generate a weekly workflow audit report?

The best way to generate a weekly workflow audit report is to run a meta-review over the last week's sessions. This checks prior reviews for overlap, runs a preprocessing script, and saves a dated Markdown report to your reviews folder.

Why does my project keep repeating the same workflow errors?

Your project keeps repeating workflow errors because recurring failures across sessions go ungrouped and unrooted. A meta-review surfaces these repeated errors, checks for existing CLAUDE.md rules that address them, and recommends edits to prevent recurrence.