agent-skills

Generate conventional commit message from staged or unstaged changes

Checks projects and packages for CVEs using Socket.dev CLI and native audit commands. Use when installing or auditing dependencies for vulnerabilities, evaluating a package before install, or scanning a project's dependency tree.

Audit the current project's memory store (`~/.claude/projects/<encoded-cwd>/memory/`) and prune memories that no longer match the project state — dead file/function/flag references, past-dated project memories, orphaned index entries, and duplicates. Suggests candidates with rationale and requires per-item confirmation before deleting; does not edit memory bodies. Use when the user says "prune memory", "clean up memory", "audit memory", "memory is stale", "remove old memories", or invokes /prune-memory.

Symlink every skill in this repo into one or more agent skills directories (defaults to ~/.claude/skills/ and ~/.agents/skills/) so the skills become globally discoverable. Use when the user says "install these skills", "make these skills global", "symlink skills to ~/.claude", "symlink skills to ~/.agents", "register the skills", or wants to update the global skill set after editing skills here.

Build, validate, and exercise the agent-skills registry. Use when asked to run agent-skills, install its deps, lint or test it, run its bats suites, exercise a Claude Code hook, regenerate the index files, or symlink the skills out to ~/.claude / ~/.agents / ~/.cursor / ~/.opencode.

Concise, technically-precise response style with kawaii flourishes (emojis, emoticons, otaku suffixes). Use when the user wants playful agent output while keeping technical accuracy and brevity.

Generates a changelog by analyzing git history and tags. Use when the user asks to generate, write, or update a changelog.

Reclaim disk space by cleaning macOS/Linux dev caches (Docker/OrbStack, package managers, Xcode, Homebrew, kilter, browser caches). Tiered by risk — safe ops run by default, destructive ops require confirmation. Use when the user mentions disk pressure, full disk warnings, "clean up", "free space", or invokes /clean-caches.

Drafts and clarifies intent and scope of a change before implementation. Use when creating user stories, tasks, plans, or other artifacts to define what to build.

Reads from an existing issue, interprets expected outcomes, plans before making changes

Refactors existing code by identifying language and applying best practices. Use when the user asks to refactor, improve, optimize, or clean up code.

Reviews the markdown and template files changed in the current session and tightens them for agent performance — sharper trigger descriptions, progressive disclosure, imperative voice, one-claim-per-bullet, fewer redundant rules — while preserving every workflow step, constraint, and trigger condition. Use when the user has just edited one or more skill or prompt files (`SKILL.md`, `references/*.md`, `recipes/*.md`, agent templates, slash-command prompts) and wants a tightening pass before committing, or invokes "/refactor-skills", "refactor this skill", "tighten the skill", "polish SKILL.md", "make this skill leaner". Behaviour-preserving only — never use this to add features, expand scope, or remove rules.

Run any long-running process in a named tmux session so it survives across Claude sessions. Use when starting any process that won't exit on its own — dev servers, clusters, watchers, builds, test runners, tail/follow commands, daemons, REPLs, port-forwards, agents, etc. — or when reading output from a previously started one, or when the user mentions tmux sessions, windows, or panes directly. The Bash tool's `run_in_background` only persists within the current Claude session; use this skill instead for anything that should outlive the session.

Analyzes web performance using Chrome DevTools MCP. Measures Core Web Vitals (LCP, INP, CLS) and supplementary metrics (FCP, TBT, Speed Index), identifies render-blocking resources, network dependency chains, layout shifts, caching issues, and accessibility gaps. Use when asked to audit, profile, debug, or optimize page load performance, Lighthouse scores, or site speed. Biases towards retrieval from current documentation over pre-trained knowledge.