skill-improver

Analyzes .claude/feedback files and updates SKILL.md with Learnings and version history.

11|1|Updated Jan 16, 2026
One-click install
npx skills add https://github.com/sunnypatneedi/claude-starter-kit --skill skill-improver-sunnypatneedi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-improver
Source: https://github.com/sunnypatneedi/claude-starter-kit/tree/main/skills/general/skill-improver
Command: npx skills add https://github.com/sunnypatneedi/claude-starter-kit --skill skill-improver-sunnypatneedi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill analyzes feedback from skill retrospectives and updates skill files to improve with use, creating a continuous improvement loop where skills get better based on real-world usage.

Core Features & Use Cases

  • Automated feedback synthesis: Collects, categorizes, and prioritizes feedback from .claude/feedback/ to drive targeted skill improvements.
  • Learnings-driven updates: Adds or refines sections like Learnings and Version History to document changes and rationale.
  • Archival and traceability: Moves processed feedback to an archive and maintains a SUMMARY log for auditing and learning.
  • Use Case: When multiple users report a missing step or confusion, the Skill Improver updates the relevant SKILL.md to improve clarity and reliability.

Quick Start

Run a quick review cycle: inspect new feedback in .claude/feedback/, identify recurring patterns, and update the affected SKILL.md with Learnings, a new version entry, and an archive note.

Frequently Asked Questions about skill-improver

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

FAQPage Schema
How do I update skill documentation based on retrospective feedback?

You can update skill documentation by scanning `.claude/feedback` files, identifying recurring patterns, and appending Learnings, version history, and documentation directly to SKILL.md files for structured skill improvement.

How does automated feedback synthesis work for skill improvement?

Automated feedback synthesis works by collecting, categorizing, and prioritizing feedback from `.claude/feedback/` directories to drive targeted updates, ensuring skills improve through a continuous improvement loop based on real-world usage.

What is the best way to maintain traceability when archiving processed feedback?

The best way to maintain traceability is to move processed feedback to an archive directory while maintaining a SUMMARY log, ensuring non-destructive changes and a clear audit trail for all SKILL.md updates.

Can I enforce non-destructive changes when updating SKILL.md files?

Yes, you can enforce non-destructive changes when updating SKILL.md files. The workflow ensures updates only add or refine sections like Learnings and version history, preserving original content while archiving processed feedback.

Does skill improvement require any specific dependencies or components?

No specific dependencies or components are required. The skill improvement workflow operates directly on `.claude/feedback` files and SKILL.md documentation without needing external libraries or frameworks.

When do I need to add a Learnings section to my skill files?

You need to add a Learnings section to your skill files when multiple users report a missing step, confusion, or recurring issues during retrospectives, prompting a structured update to improve clarity and reliability.