claude-code

This skill should be used when the user asks to "check clean code", "review code cleanliness", "find code smells", "is this function too long", "are these names clear", "review my error handling", "check my tests", or "clean up this code". Also triggers when the user mentions Clean Code, Uncle Bob, Robert Martin, meaningful names, function length, comment rot, F.I.R.S.T. tests, or names a dimension directly ("naming", "boundaries", "concurrency"). Supports auditing all nine dimensions at once or focusing on a single dimension.

Scan Claude Code chat session logs for recurring agent anti-patterns (hedging, stalling, skipping, false completions, preference-asking, dismissing, cosmetic, caveats, flagging) and produce a ranked report with frequency counts, example excerpts, and suggested guardrail rules. Use when user asks to "analyze sessions for anti-patterns", "find bad patterns in logs", "what anti-patterns am I seeing", "agent-guardrails analyze", or wants data-driven guardrail recommendations.

Install agent behavioral guardrail rules into a project's .claude/ directory. Generates a Stop hook bash script and configures settings.local.json. Includes nine battle-tested rules (no-guessing, no-stalling, no-preference-asking, no-false-completion, no-skipping, no-dismissing, no-cosmetic, no-caveats, no-flagging). Use when user asks to "install guardrails", "set up agent guardrails", "add behavioral hooks", "agent-guardrails install", or wants to enforce assistant discipline.

Test installed agent guardrail hooks by sending trigger phrases through the stop hook script. Checks hook installation first, then runs two test phrases per rule. Use when user asks to "test guardrails", "verify hooks", "check guardrail patterns", "agent-guardrails test", or wants to confirm hooks are working.

Re-analyze Claude Code session logs against existing agent guardrail rules to measure effectiveness, catch false positives, identify missed anti-patterns, and refine regex patterns. Use when user asks to "update guardrails", "refine guardrail rules", "check guardrail effectiveness", "agent-guardrails update", "tune guardrails", or wants to improve existing behavioral rules based on real usage data.

Use when the user wants to write a "product spec" or "product specification", frame a "product bet", or define a falsifiable hypothesis before building. Also triggers on "replace the PRD", "retire the PRD", "the PRD is too long / too vague", "tighten this spec for an engineer or agent", "what's the bet", "what problem are we solving", "acceptance criteria", "success criteria", "kill / scale / graduate thresholds", "measurement plan", or when scoping a feature's problem, bet, success behaviors, and evaluation before any behavioral spec. Part of the spec-dd workflow; precedes /spec-dd:spec.

This skill should be used when the user asks to "write specifications", "create test specifications", "specification-driven development", "spec-first", "behavioral specs", "derive test scenarios", "test implementation specification", "check specification alignment", "review specs", "verify implementation", or "spec-dd". Also triggers when the user mentions "SDD", "SDD-TDD", "spec-driven", "behavioral testing workflow", "test-first design", or asks about writing specifications before code, deriving tests from specs, or verifying implementation against specifications. Supports a full workflow walkthrough or focusing on individual phases.

Generate arc42 architecture documentation from a codebase. Read the repository into a structured evidence base, then author the 12 arc42 sections — emitting Mermaid diagrams where the code gives high-confidence structure, and explicit typed GAP flags where human input is needed, never fabricating. Use when asked to create, scaffold, or update arc42 docs, an architecture document, or per-section architecture views (building block view, runtime view, deployment view, etc.) for a project.

Draw ASCII-art / text diagrams for software documentation using extended ASCII box-drawing characters (IBM CP437 codes > 128 — single-line ┌─┐│└┘├┤┬┴┼, double-line ╔═╗║╚╝, shading ░▒▓█), never +, -, |, or /. Enforces ≤80-char width and strict row/column alignment, renders in monospaced fonts. Use when the user wants an ascii diagram, a text/box diagram, a terminal diagram, or a plain-text visualization.

Military-style communication mode — enforce extreme brevity, lead with critical issues and end with a one-line bottom-line conclusion, bullets over prose, and cut filler, preambles, code examples, and hand-holding. Use when the user types "tldr", "/tldr", "short mode", "be brief", or "concise mode", or otherwise asks for terse, no-fluff output.

Router for the diagrams plugin — pick the right diagram format and dispatch to its skill. Routes to ascii-art (text/box-drawing diagrams), mermaid (flowcharts, sequence, class, state, ER, gantt), or wardley (Wardley maps in markdown). Use when the user says "draw a diagram", "diagram this", "make a chart", "visualize this as a diagram", "ascii diagram", "mermaid diagram", or "wardley map".

Produce a Mermaid diagram (flowchart, sequence, class, state, ER, gantt, pie, mindmap) inside a ```mermaid fenced code block for rendering in Markdown, docs, or the GitHub/GitLab viewer. Use when the user asks for a mermaid diagram, a flowchart/sequence/class/state/ER/gantt diagram, or a diagram to embed in Markdown.

Produce a Wardley map in the WTG2 DSL (wardleyToGo) as a .wtg2 file that renders to SVG via wtg2svg — anchors (user needs), components positioned by value chain (visibility) and evolution (roman numerals I–IV), dependencies, movement, inertia, build/buy/outsource, signals, gameplays, and groups. Use when the user asks for a Wardley map, strategic map, or value-chain map.

Arm the current Claude Code session to self-terminate by uttering a kill phrase once all work is finished and no questions remain. Use when the user says "terminate when done", "whendone", "end the session when finished", or "self-terminate when complete".

Remove the terminator kill hooks from local or global scope. Strips single-kill.sh and double-kill.sh Stop hook entries from the appropriate settings file (preserving other hooks), and deletes the hook scripts and config. Use when the user asks to "remove terminator", "uninstall the kill hooks", or "disable session termination".

Show the configured kill phrases and case sensitivity for both local (project) and global (user) terminator installations. Use when the user asks "what is my kill phrase", "show terminator config", "terminator info", or "what phrase is configured".

Install the terminator kill hooks at project (local) or user (global) scope. Installs single-kill.sh and/or double-kill.sh as Stop hooks, each bound to a phrase, merging into the appropriate settings file without clobbering existing hooks. Use when the user asks to "install terminator", "set up the kill hooks", "add single kill", "add double kill", or "enable session termination".

Router for the terminator plugin — Stop hooks that end a Claude Code session (and optionally its terminal) when the agent's final message contains a kill phrase. Routes to install, remove, update, or whendone. Use when the user says "terminator", "install the kill hook", "set up a kill phrase", "single kill", "double kill", "remove the terminator hook", "change the kill phrase", or "terminate when done".

Update the terminator kill phrases or case sensitivity at local or global scope. Prints the current phrase before asking for a new one. Rewrites only the changed fields; hooks read config live so changes take effect immediately. Use when the user asks to "update the kill phrase", "change the single/double kill phrase", or "toggle case sensitivity".

This skill should be used when the user asks to "analyze documentation structure", "audit progressive disclosure", "map documentation references", "find orphaned docs", "check context rot risk", "audit CLAUDE.md", "audit AGENTS.md", or wants to understand how a repository's documentation is structured for AI agent consumption. Produces a disclosure map, orphan report, and anti-pattern findings.

This skill should be used when the user asks to "restructure CLAUDE.md", "restructure AGENTS.md", "create a documentation index", "add a table of contents to CLAUDE.md", "organize project docs", "refactor root configuration file", "build a doc index", or wants to restructure how a repository's root configuration file references its documentation. Generates a thematic, book-style index in the highest-precedence root configuration file.

Use when the user wants to initialize, build a collection in, remove a collection from, clear, or inspect the `guard` CLI file-permission tool. Triggers on "init guard", "set up guard", "guard my test files", "create a guard collection", "remove guard collection", "delete guard collection", "unguard a collection", "clear all guard", "wipe guard", "reset guard", "guard status", "guard info", "show guard", "what's guarded".

This skill should be used when the user asks to "update the changelog", "generate a changelog", "create CHANGELOG.md", "write release notes", "update CHANGELOG", "add changelog entry", "changelog from commits", or mentions changelog generation, release documentation, or keeping a changelog up to date. Auto-detects whether to create or update based on file existence. Follows Keep a Changelog format with Semantic Versioning.

This skill should be used when the user asks to "create a changelog", "initialize CHANGELOG.md", "start a changelog", or when the main changelog skill detects no existing CHANGELOG.md file. Creates a new CHANGELOG.md from scratch by analyzing the full git commit history and organizing changes by date-based version sections.

This skill should be used when the user asks to "update the changelog", "add to the changelog", "changelog for new release", "update CHANGELOG.md", or when the main changelog skill detects an existing CHANGELOG.md file. Appends new version entries by analyzing commits since the last documented version.

This skill should be used when the user asks to "check SOLID violations", "audit class design", "review code quality", "find design smells", or "improve object-oriented architecture". Also triggers when the user mentions a principle by name (e.g., "check SRP", "is this violating LSP?", "dependency inversion"), or asks about coupling, cohesion, or class responsibilities. Supports checking all five principles at once or focusing on a single principle.

This skill should be used when the user asks to "analyze documentation structure", "audit progressive disclosure", "restructure CLAUDE.md", "restructure AGENTS.md", "create a documentation index", "organize project docs", "find orphaned docs", "map documentation references", "improve context loading", or mentions progressive disclosure, documentation hierarchy, root configuration file organization, or context rot prevention. Analyzes or restructures how a repository discloses documentation through root configuration files.

This skill should be used when the user asks to "check architecture principles", "audit system design", "review code for coupling", "find architecture smells", or "improve system-level design". Also triggers when the user mentions a principle by name (e.g., "check separation of concerns", "is this violating DRY?", "Law of Demeter", "KISS", "YAGNI", "resilience", "evolvability", "loose coupling"). Supports checking all ten principles at once or focusing on a single principle.

This skill should be used when the user wants to "hand off", "create a handoff", "wrap up for the next session", "pass this to another agent", "continue from a handoff", "pick up where I left off", "resume a previous session", "load a handoff", or mentions handoff documents, session continuity, or task continuation.

Use when you have a specified engineering task or task list and need to decide which AI model should handle each work unit — routes tasks across Claude Opus/Sonnet/Haiku, GPT-5.4 family, GPT-5.3-Codex, and Gemini 2.5/3 with per-unit category, evidence-based rationale, context briefing, and failure-mode warnings. Also checks whether the task is broken down enough and proposes finer decomposition before routing.

This skill should be used when the user wants to "create a claudeignore", "update claudeignore", "optimize context", "ignore folders", "reduce token usage", "set up .claudeignore", or mentions .claudeignore, context window optimization, or ignoring directories from Claude Code indexing.

Analyze a repository and generate a new .claudeignore file that excludes directories and files wasting context tokens.

Update an existing .claudeignore file by adding, removing, or revising entries based on current repository state.

Find handoff documents and continue work on a previous task with full context. Companion to the `handoff` skill.

Compact the current conversation into a structured handoff document so a fresh agent (or human) can continue the work with zero prior context.

This skill should be used when the user asks to "check for access control issues", "analyze authorization", "find IDOR vulnerabilities", "audit CORS configuration", "check for privilege escalation", or mentions "access control", "authorization", "IDOR", "CORS", "JWT tampering", or "directory traversal" in a security context. Maps to OWASP Top 10 2021 A01: Broken Access Control.

This skill should be used when the user asks to "check API security", "audit REST API", "find BOLA vulnerabilities", "check for mass assignment", "analyze API rate limiting", "detect excessive data exposure", or mentions "API security", "BOLA", "IDOR", "mass assignment", "rate limiting", "broken function-level authorization", "excessive data exposure", or "OWASP API Top 10" in a security context.

This skill should be used when the user asks to "map attack surface", "list entry points", "inventory API endpoints", "find all inputs", "enumerate routes", "discover exposed endpoints", or "map external interfaces". Also triggers when the user asks about exposed APIs, form handlers, file upload endpoints, WebSocket handlers, CLI argument parsers, or wants to understand where external data enters the system.

This skill should be used when the user asks to "check for authentication issues", "analyze auth", "find credential vulnerabilities", "review login security", "check session management", or mentions "authentication", "passwords", "MFA", "sessions", or "brute force" in a security context. Maps to OWASP Top 10 2021 A07: Identification and Authentication Failures.

This skill should be used when the user asks to "check business logic security", "find logic flaws", "audit workflow security", "check for coupon abuse", "detect negative amount exploits", "analyze state machine security", or mentions "business logic", "workflow bypass", "negative amount", "coupon abuse", "self-referral", "state manipulation", or "time-based exploit" in a security context.