Broad code reviewer combining confidence scoring, knowledge preservation, and architectural analysis across 5 categories
Reviews code for knowledge preservation — undocumented decisions, implicit assumptions, comprehension risks, and lost context that makes code unmaintainable
Fast generalist code reviewer for bugs, logic errors, security patterns, code quality, and project convention violations. Sonnet-powered for speed; complements deeper specialized reviewers.
Cross-file structural analyst for refactoring opportunities — duplication, naming consistency, validation scattering, interface drift, module boundaries, zombie code.
Assess severity of code coverage gaps and suggest targeted test improvements
Automated review-fix-recheck loop. Runs code review, triages findings, fixes autonomously via subagent, and re-reviews if substantial changes were made. Iterates until code stabilizes.
Run parallel code review agents and consolidate findings into a unified report
Run code coverage analysis and review gaps for severity and test suggestions
Walk through coverage review findings interactively, writing tests for gaps
Walk through code review findings interactively, addressing each issue. Use "auto" for autonomous fixing with TDD.
Uses power tools
Uses Bash, Write, or Edit tools
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Personalized Claude Code plugins with specialized agents, skills, and conventions for development and code review.
Originally forked from everything-claude-code. You probably want that one rather than this one that has been tailored to my own personal workflow.
This repo provides four plugins that can be installed independently:
| Plugin | Description |
|---|---|
decaf | General-purpose agents and skills (analysis, development) |
decaf-review | Multi-agent code review, coverage analysis, and refactoring |
decaf-planning | Planning skills for PRDs, implementation plans, and phase breakdowns |
decaf-memory | Memory skills backed by erinra (hybrid semantic search, knowledge graphs) |
decaf-experimental | Experimental skills being tested before promotion to core |
# Add as a marketplace
/plugin marketplace add alphaleonis/decaf-claude-config
# Install plugins
/plugin install decaf-claude-config@decaf
/plugin install decaf-claude-config@decaf-review
/plugin install decaf-claude-config@decaf-planning
/plugin install decaf-claude-config@decaf-memory
/plugin install decaf-claude-config@decaf-experimental
Or install from a local clone:
cd /path/to/decaf-claude-config
/plugin marketplace add ./
/plugin install decaf-claude-config@decaf
/plugin install decaf-claude-config@decaf-review
/plugin install decaf-claude-config@decaf-planning
/plugin install decaf-claude-config@decaf-memory
/plugin install decaf-claude-config@decaf-experimental
Memory skills require the erinra MCP server. See the erinra repo for installation options.
# Install erinra — see https://github.com/alphaleonis/erinra-mcp#installation
# Register the MCP server
claude mcp add erinra -- erinra serve -s user
# Install the plugin
/plugin install decaf-claude-config@decaf-memory
decaf-claude-config/
├── .claude-plugin/ # Marketplace manifest
├── conventions/ # Shared convention files (@file references)
├── decaf/ # Core plugin
│ ├── .claude-plugin/plugin.json
│ ├── agents/ # 6 agents
│ └── skills/ # 5 skills
├── decaf-review/ # Review plugin
│ ├── .claude-plugin/plugin.json
│ ├── agents/ # 10 agents
│ └── skills/ # 6 skills
├── decaf-planning/ # Planning plugin
│ ├── .claude-plugin/plugin.json
│ └── skills/ # 6 skills
├── decaf-memory/ # Memory plugin (erinra)
│ ├── .claude-plugin/plugin.json
│ └── skills/ # 4 skills
├── decaf-experimental/ # Experimental plugin
│ ├── .claude-plugin/plugin.json
│ └── skills/ # 4 skills
├── CLAUDE.md
└── README.md
decaf — Core SkillsSkills are invoked as /decaf:<skill-name>.
| Skill | Description |
|---|---|
decision-critic | Stress-test decisions through adversarial analysis |
incoherence-detector | Detect doc/code/spec inconsistencies |
note | Capture a follow-up task as a bean without interrupting current work |
powershell-expert | PowerShell development patterns |
problem-analysis | Root cause investigation |
decaf — Core AgentsAgents are referenced via the Task tool as decaf:<agent-name>.
| Agent | Purpose |
|---|---|
architect | Feature architecture design |
csharp-developer | C# implementation with idiomatic patterns |
go-developer | Go implementation with idiomatic patterns |
debugger | Systematic debugging with evidence gathering |
planner | Implementation planning |
technical-writer | LLM-optimized documentation |
decaf-review — Review SkillsSkills are invoked as /decaf-review:<skill-name>.
| Skill | Description |
|---|---|
code-review | Parallel multi-agent code review with consolidation |
coverage-review | Run code coverage analysis and review gaps for severity |
refactor | Analyze code for structural improvement opportunities and produce a prioritized plan |
handle-cr | Walk through code review findings interactively |
handle-coverage | Walk through coverage gaps interactively, writing tests |
handle-refactoring | Walk through refactoring opportunities interactively |
decaf-review — Review AgentsAgents are referenced via the Task tool as decaf-review:<agent-name>.
Memory skills backed by erinra memory MCP server
Planning skills for PRDs, implementation plans, and phase breakdowns
General-purpose agents and skills for Claude Code
npx claudepluginhub alphaleonis/decaf-claude-config --plugin decaf-reviewReflective code review toolkit with structured analysis, blast radius analysis, domain expertise, and intelligent review orchestration
Self-review workflow: Simplifier + Scrutinizer for code quality
使用多個專門代理進行自動化程式碼審查,配備基於置信度的評分系統以過濾誤報
Automated code review with severity levels and actionable feedback
Complete collection of battle-tested Claude Code configs from an Anthropic hackathon winner - agents, skills, hooks, and rules evolved over 10+ months of intensive daily use
Complete creative writing suite with 10 specialized agents covering the full writing process: research gathering, character development, story architecture, world-building, dialogue coaching, editing/review, outlining, content strategy, believability auditing, and prose style/voice analysis. Includes genre-specific guides, templates, and quality checklists.