Agent Stuff

Personal prompts, skills, and commands I use with Claude and Cursor. Some are highly specific to my own setup (like my Obsidian vault). Sharing publicly in case any of it's useful or interesting to others.

What's here

• Skills — Instructions that Claude/Cursor automatically apply based on context
• Commands — User-invoked slash commands for specific workflows
• Reference — Documentation on Agent Skills, Cursor rules, and how components work together
• Templates — Starting points for creating new skills, commands, and rules

Agent skills format

This repo uses the Claude Agent Skills format. Skills are folders containing a SKILL.md with YAML frontmatter and instructions. The description field tells AI agents when to load and apply the skill.

skill-name/
├── SKILL.md              # Metadata + instructions
├── reference.md          # Optional: detailed docs
└── supporting-file.md    # Optional: additional context

Commands are standalone markdown files with prompts that get invoked via /command-name.

Using these

Copy whatever looks useful into your own .claude/skills/ or .cursor/skills/ directory. Adapt as needed.

You can also use the skills CLI to quickly add items from this repo:

bunx skills add nweii/agent-stuff

Or use npx if you prefer.

Note: Some skills reference personal vault notes not included in this repo.

License

MIT — do whatever you want with it.

---

Catalog

Auto-generated by scripts/update-catalog.py

Skills

dev

• commit — Create well-formatted commits with conventional commit messages. Use when user asks to commit, wants to commit changes, or needs help with commit messages.
• raycast-extensions — Build Raycast extensions using React and Node.js. Use for creating commands, UI components (List, Form, Grid), hooks (useFetch), AI integration, and manifest configuration.
• react-useeffect — React useEffect best practices. Use when writing Effects, derived values, or data fetching. Teaches when NOT to use Effects and better alternatives like useMemo or key props.
• rename-files — Analyze file content and propose intelligent renames using context-aware naming conventions. Date-prefixed for transactional/periodic documents, content-first for creative works. Use for organizing files, cleaning up downloads, or standardizing filenames.
• spec-shaping — Shape product ideas into actionable specs and sprint plans. Use when interviewing about a product plan, breaking specs into sprints, or turning vague ideas into well-scoped work.
• update-changelog — Update changelog files following Keep a Changelog conventions. Use for project changelogs or skill changelogs. Handles version sections, change type groupings, and skill-specific metadata updates.
• use-tailwind-v4 — Tailwind CSS v4 reference and migration guide. Use for v4 projects, syntax changes, upgrading from v3, and troubleshooting v4-specific utility patterns or configuration differences.
• using-heavy-mcps — Use mcporter and jq to reduce token bloat from heavy MCPs like Sanity or Brain vault. Filter and compact tool outputs outside chat to feed only essential data back to the model.
• web-design-guidelines — Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".

general

• archive-conversation — Create analytical archival summaries of AI conversations, capturing intellectual journeys, key insights, and technical logs. Use when archiving, saving, or documenting a chat session.
• find-fallacies — Analyze text for logical fallacies. Use when reviewing arguments, debates, articles, or reasoning that may contain flawed logic.
• lookup — Explain the meaning of selected text using surrounding context. Identifies unclear words, phrases, idioms, or colloquialisms and provides concise definitions with cultural context and usage notes.
• suggest-lucide-icons — Suggest relevant Lucide icons for concepts or UI placements. Searches lucide.dev to find actual icons that symbolize ideas or fit design contexts. Use when needing icons for UI work, documentation, or symbolic representation.
• visual-keywords — Generate keyword-rich strings for visual references to optimize for fuzzy search and recall. Focuses on searchability and keyword matching rather than prose or alt text.

personal

• aid-finances — Expert financial guidance for budgeting, investments, and retirement planning. Use for analyzing market strategies, debt management, or general personal finance questions.
• culinary-assistant — Culinary guidance for cooking, recipes, and meal planning. Use for kitchen techniques, substitutions, and format conversion (Mela, Schema.org). Handles recipe parsing and improvements.
• sync-things-tasks — Syncs tasks between Obsidian vault and Things 3. Adds tasks from notes, extracts action items from projects, reviews existing tasks. Use when managing todos from vault content.
• things-mac — Manage Things 3 on macOS via CLI. Read, search, add, or update tasks, projects, and areas. Use for task management, listing inbox/today, and inspecting tags or repeating templates.
• vault-synthesis — Synthesize information across vault notes: history rollups, frontmatter summaries, periodic rollups, and topic note creation. Use when aggregating, summarizing, or organizing content across multiple notes.

strategy

• validate-startup-ideas — Find and validate startup ideas by mining user complaints, crafting premises, and navigating the idea maze. Use when discovering product opportunities, validating ideas, shaping solutions, researching user pain points, or exploring what to build.

Commands

general

• cleanup-dictation — Process dictated or stream-of-consciousness input into clean formatted output. Used for Superwhisper presets and similar dictation workflows.
• create-command — Guide for creating new custom Claude commands with proper structure.