improve-codebase-architecture
Identify structural codebase issues and propose refactoring directions
Source: mattpocock
At a glance
- Best for
- Investigate structural issues using real code
- Works with
- The codebase and areas of concern · GLOSSARY.md and existing ADRs, if available
- Outcome
- Architecture-issue candidates and an HTML report · A clear refactoring direction
- Source
- mattpocock/skills
01 / Understand the skill
About improve-codebase-architecture
Read domain terminology and architecture decisions, examine module interfaces and friction in making changes, and present candidates in an HTML report. Use follow-up questions to clarify the refactoring direction and maintain domain documents. Requires codebase-design, grilling, domain-modeling, and HTML report templates to assess interfaces, scope, and migration constraints in a real project.
Requirements
- An agent with codebase access and subagent capabilities
- The companion codebase-design, grilling, and domain-modeling skills
- Complete reference files, including HTML-REPORT.md
Inputs
- The codebase and areas of concern
- GLOSSARY.md and existing ADRs, if available
Outputs
- Architecture-issue candidates and an HTML report
- A clear refactoring direction
- Terminology or ADR updates based on the discussion
02 / Confirm the fit
Use improve-codebase-architecture when you need to
- Investigate structural issues using real code
- An HTML report of candidate approaches
- Incorporate domain terminology and architecture decisions
03 / See the workflow
How improve-codebase-architecture works
- 01Read the code, domain terminology, and architecture decisions
- 02Evaluate and present improvement candidates
- 03Choose a direction, then clarify interfaces and migration constraints
Check before installing
Compatible platforms
Cursor · GitHub Copilot · OpenAI Codex · Claude Code
Best fit
Investigate structural issues using real code
Review the source instructions if the required output differs from the formats described above.
Install and use
Compatible with Cursor and GitHub Copilot and OpenAI Codex and Claude Code · Published by mattpocock
Ready to add improve-codebase-architecture?
Use the published instructions below, then provide the required input and describe the result you need.
General instructions · Cursor · GitHub Copilot · OpenAI Codex · Claude Code
claude plugin install mattpocock-skills@claude-plugins-official Install the complete plugin as described in the upstream README to preserve cross-Skill calls. The Codex entry point is also recorded in the README Raw snapshot.
Review the source repository and license before installing third-party skills.
Continue exploring
Similar skills
Authoritative reference for the Claude API and Anthropic SDKs: model IDs, pricing, thinking/effort params, streaming, tool use, caching, managed agents, and model migration.
Read, edit, create, and fix Excel/CSV files — formulas, formatting, charts, and clean financial models with zero formula errors.
Generate distinctive, production-grade frontend UI that avoids generic AI aesthetics — pages, components, dashboards, and landing pages.
Build high-quality MCP (Model Context Protocol) servers in Python (FastMCP) or Node/TypeScript that let LLMs use external services through well-designed tools.
