This skill should be used when the user asks to "check adherence", "check rule adherence", "are my rules being followed", "is Claude following my instructions", "how effective is my CLAUDE.md", "rule compliance", "instruction effectiveness", "how are my rules performing", "which rules are being violated", "audit my rules", "check my CLAUDE.md", mentions "/check-rules", or wants to know whether their coding agent instruction rules are actually being followed across Claude Code sessions.
Use when the user wants to infer reusable project conventions from existing code and turn them into instruction rules: "discover conventions", "infer rules from code", "what rules should I add", "generate rules from this codebase", or "/discover". This is for convention mining, not for checking rule adherence or explaining how a specific module works.
This skill should be used when the user asks "has this module drifted", "is this code still doing what it was designed for", "scope creep analysis", "architectural drift", "what was this originally for", "how has this changed over time", mentions "/drift", or wants to detect when code has quietly evolved away from its original purpose.
Use when the user wants a deep explanation of a specific file, module, or implementation flow: "explain this code", "walk me through this file", "why is this structured this way", or "/explain". Best for focused code understanding with history context, not for repo-wide architecture mapping, recent-change summaries, or generic "what happened?" requests.
This skill should be used when the user asks to "find hotspots", "what are the riskiest files", "where do bugs hide", "churn analysis", "which files change most", "find risky code", "technical debt hotspots", mentions "/hotspots", or wants to identify the highest-risk areas of their codebase based on change frequency, complexity, and coupling.
This skill should be used when the user asks to "check the impact", "what will break if I change this", "blast radius", "impact analysis", "what depends on this", "who imports this", "is it safe to change this", mentions "/impact", or wants to understand the consequences of modifying a specific file or function before making changes.
This skill should be used when the user asks to "lint instructions", "lint my CLAUDE.md", "check instruction quality", "review my rules", "analyze my CLAUDE.md", "find issues in my instructions", "improve my rules", "instruction quality", mentions "/lint-rules", or wants to check the quality of their coding agent instruction files (CLAUDE.md, .cursorrules, AGENTS.md, .claude/rules/, .claude/agents/, .claude/skills/).
Run for repo-wide or diff-wide readiness sweeps: "/orbit", "/orbit pr", "/orbit deep", "/orbit quick", "audit this project", "project-wide review", "is this PR ready overall", or "combined security, tests, codebase, evolution, and instruction review". Use this for broad multi-system assessment, not for a single-file review or a one-domain question.