sus
Finds suspicious, architecturally problematic, or high-impact maintainability issues in a codebase. Deploys parallel analysis agents to explore code, then synthesizes findings into a prioritized report. Use when user says "find problems", "audit code", "what's sus", "code review the repo", "find tech debt", or asks about code quality.
explore
Strategic discovery of a project's capabilities from a solutions architect perspective. Deploys parallel discovery agents to map architecture, inventory features, and assess infrastructure, then synthesizes findings into a capabilities report with strategic improvement recommendations. Use when user says "explore", "what does this do", "project overview", "capabilities", "feature inventory", or asks about strategic direction.
init-repo
Scaffolds a new project repository for agentic development from a problem description. Guides the user through naming, runtime, package manager, devops, licensing, and project structure, then initializes git and a CLAUDE.md file. Use when the user says "init project", "new project", "start a project", "scaffold repo", "create repo", or describes a problem they want to build a solution for.
combine-docs
Discovers all documentation in a project, analyzes it for redundancy, contradictions, and staleness, proposes a consolidation plan, then executes it. Produces optimized docs for both humans and agents. Use when user says "combine docs", "consolidate documentation", "merge docs", "clean up docs", "fix documentation", "optimize docs", "documentation sprawl", "too many docs", or asks about doc organization.
check
Verifies the agent's current work against a specific question by analyzing unstaged changes, staged changes, recent commits, and codebase context. Answers succinctly for a senior audience. Use when user says "/check", "verify that", "confirm that", "check if", "is X done?", or asks about current session changes.