improve-codebase

Decompose and audit a repository into reviewable chunks with iterative fixes.

9|1|Updated Jan 8, 2026
One-click install
npx skills add https://github.com/conn-castle/agent-layer --skill improve-codebase
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: improve-codebase
Source: https://github.com/conn-castle/agent-layer/tree/main/internal/templates/skills/improve-codebase
Command: npx skills add https://github.com/conn-castle/agent-layer --skill improve-codebase

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Deep, autonomous audit and improvement of the entire codebase (or user-specified subsystems): survey the repository structure, decompose into reviewable chunks, run parallel multi-lens reviews, fix findings iteratively, and delegate to complementary skills (simplify-code, boost-coverage, audit-tests, fix-issues) where appropriate. Use after a release, during a maintenance window, or whenever a thorough independent quality sweep is needed.

Core Features & Use Cases

  • Systematic survey that maps the repository structure into reviewable chunks
  • Parallel, multi-lens auditing across chunks with architecture, correctness, and quality lenses
  • Iterative fixing and delegation to complementary skills, with a master report and ISSUES.md log
  • Strong governance: pre-defined artifacts, run-id workflow, and cross-cutting review

Quick Start

Run the improve-codebase skill to perform an end-to-end repository health audit and implement fixes across chunks.

Frequently Asked Questions about improve-codebase

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

FAQPage Schema
How do I automate a full repository audit and fix codebase quality issues?

Automate a full repository audit by surveying the repository structure, decomposing it into reviewable chunks, and running parallel multi-lens reviews to iteratively fix findings. This delivers a master report, per-chunk findings, and an ISSUES.md log.

What is the best way to perform a systematic codebase review across multiple sub-systems?

A systematic codebase review maps the repository into chunks, applies architecture, correctness, and quality lenses, and coordinates with complementary skills. This ensures cross-cutting analysis and iterative fixes across all sub-systems.

Can I run an automated codebase improvement workflow during a maintenance window?

Yes, you can run an automated codebase improvement workflow during maintenance windows or after a release. It performs an end-to-end quality sweep, generating governance artifacts and coordinating sub-skills to resolve findings.

Does the automated repository audit support chunked reviews for large codebases?

The automated repository audit supports chunked reviews by decomposing the codebase into manageable sections. This enables parallel multi-lens auditing and iterative fixing across modules and sub-systems.

How do codebase improvement workflows handle cross-cutting analysis and issue tracking?

Codebase improvement workflows handle cross-cutting analysis by applying multiple review lenses to decomposed chunks and tracking all findings in an ISSUES.md log, ensuring iterative fixes are documented in a final master report.