rz-self-improve

Scan transcripts and repository files for inconsistencies and propose fixes.

1|Updated Apr 8, 2026
One-click install
npx skills add https://github.com/rczamor/rz-claude-skills --skill rz-self-improve
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: rz-self-improve
Source: https://github.com/rczamor/rz-claude-skills/tree/main/plugins/rz-skills/skills/rz-self-improve
Command: npx skills add https://github.com/rczamor/rz-claude-skills --skill rz-self-improve

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve?

This Skill automates the process of refining the corpus and skills in the rz-claude-skills repository, ensuring consistency and quality.

Core Features & Use Cases

  • Retrospective Mode: Analyzes session transcripts for user corrections, contradicted facts, and repeated proof points, then proposes changes.
  • Curate Mode: Inspects the repository for broken cross-references, duplicate content, drift, and schema violations, then proposes fixes.
  • Conservative Approach: Always creates draft PRs and applies strict validation gates to prevent false positives and merges.

Quick Start

Run the 'rz-self-improve' skill with the 'retrospective' or 'curate' mode, followed by any relevant parameters.

Frequently Asked Questions about rz-self-improve

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

FAQPage Schema
How do I automate repository refinement and fix inconsistencies in my code corpus?

Automating repository refinement involves scanning session transcripts and repository files to detect inconsistencies, broken cross-references, and duplicate content, then proposing fixes through draft pull requests.

What is retrospective skill improvement for code quality?

Retrospective skill improvement analyzes session transcripts to extract user corrections and contradicted facts, using these insights to propose targeted changes that enhance overall code corpus consistency.

How do I find and fix broken cross-references and schema violations in a repository?

Finding and fixing broken cross-references and schema violations requires inspecting the repository structure to detect drift and duplicates, then generating proposed fixes validated by strict quality gates.

Can I safely automate code corpus maintenance without merging false positives?

Automating code corpus maintenance safely requires a conservative approach that creates draft pull requests and applies strict validation gates, preventing false positives from being merged directly into the main branch.

Does automated skill refinement require special permissions for repository data?

Automated skill refinement requires both read access to scan repository data and transcripts, and manipulation permissions to create draft pull requests proposing the necessary corpus and skill corrections.

What's the best way to curate a repository for duplicate content and drift?

Curating a repository for duplicate content and drift is best handled by running an automated inspection mode that scans files for schema violations and inconsistencies, then proposes validated fixes via draft PRs.