claude-universe
Systematic code review, testing, security, and architecture audits
All Skills in This Repository (20)
Pure Emerald Level Indicatorsforecast
Analyzes git history to forecast file growth trends and risks.
dissect
Trace git history to identify when and why file or function complexity was introduced.
recap
Summarize git repository activity into a structured JSON recap file.
threat-model
Generate STRIDE-based threat models for software features and modules.
orbit
Audit codebases for security, tests, code quality, evolution, and instruction violations.
discover
Detect hidden code conventions and generate evidence-backed rule suggestions.
explain
Generate context-aware code explanations from file content and git history.
trace
Traces feature execution paths with file-and-line step-by-step flows.
test
Generate unit and integration tests with deep assertions across JavaScript, Python, and Go.
test-review
Review JavaScript/TypeScript test files and produce a graded quality scorecard.
check-rules
Check rule compliance in codebases and Claude sessions using CLAUDE.md files.
hotspots
Identify high-risk files in a codebase using churn, complexity, and coupling.
Frequently Asked Questions
FAQPage SchemaHow to install claude-universe?▼
Run `npx skills add mbwsims/claude-universe --all -g -y` in your terminal to install all 20 commands globally.
How to check if a PR is ready to merge?▼
Run the /orbit pr command to get a diff-aware readiness review covering security, test coverage, code quality, and rule conformance for only your changed files.
Can it write tests for my code automatically?▼
Yes. The /test command detects your test framework, analyzes the code's input space, and writes deep-assertion tests that catch real bugs, then runs them to verify.
How to find security vulnerabilities in my code?▼
Use /scan for project-wide vulnerability detection or /security-review for an attacker-minded exploitability review of specific files, including data flow tracing and concrete fixes.
Does claude-universe work without MCP servers?▼
Yes. Every command falls back to manual analysis using file reading and git history, though the bundled MCP servers add deterministic scoring and batch analysis for larger codebases.
Related Repositories in Software Engineering
View All in Software Engineering→openclaw
Run a personal AI assistant across your devices and chat apps
superpowers
Gives coding agents a disciplined workflow from idea to merged code
react
AI agent skills for building, testing, and porting React core