skills
stuffbucket/skills/llms.txt
AI assistant skills for Claude, Copilot, and other LLMs, following the Agent Skills specification. Include use stuffbucket in a prompt to force-route through listskills before answering. Skills also auto-trigger from "Use when..." sentences in each description. Recommended priming for system prompts: instruct the agent to call listskills early for any task that might match a packaged workflow, load matches via get_skill, and follow the skill's "Use when..." guidance when present. npm run setup # create MCP symlinks + build skill…
llms.txt4 starsChanged 7 months ago
- Reads credentials
# Agent Skills AI assistant skills for Claude, Copilot, and other LLMs, following the Agent Skills specification. ## Trigger phrase Include `use stuffbucket` in a prompt to force-route through `list_skills` before answering. Skills also auto-trigger from "Use when..." sentences in each `description`. Recommended priming for system prompts: instruct the agent to call `list_skills` early for any task that might match a packaged workflow, load matches via `get_skill`, and follow the skill's "Use when..." guidance when present. ## Quick Start npm run setup # create MCP symlinks + build skill index npm run build:index # rebuild skill index only npm run build:llms # regenerate llms.txt from template npm run lint # run all linters (markdown, JS, JSON schemas) npm run validate # validate example skill frontmatter npm run test # lint + validate + build index ## Repository Health Skill validation: ✓ azure-cli-setup ✓ boundary ✓ boundary-domain-closure ✓ boundary-drift-detection ✓ boundary-generation-control ✓ boundary-noise-model ✓ boundary-scope-escape ✓ boundary-seed-encoding ✓ code-analysis-skill ✓ code-review-cycle ✓ colima-docker-setup ✓ design ✓ design-adapt ✓ design-animate ✓ design-arrange ✓ design-audit ✓ design-bolder ✓ design-check ✓ design-clarify ✓ design-colorize ✓ design-context ✓ design-critique ✓ design-delight ✓ design-distill ✓ design-extract ✓ design-frontend ✓ design-harden ✓ design-normalize ✓ design-onboard ✓ design-optimize ✓ design-overdrive ✓ design-polish ✓ design-quieter ✓ design-taste ✓ design-typeset ✓ design-typography-rules ✓ example-skill ✓ figma-make-to-vite ✓ file-management-skill ✓ ghostty-config ✓ git-workflow-skill ✓ npm-trusted-publishing ✓ pages ✓ pages-build-vite ✓ pages-commit-vite ✓ pages-prepare-vite ✓ pages-publish-vite ✓ pages-push-vite ✓ react ✓ react-best-practices ✓ react-composition ✓ skill-creator ✓ skill-eval-loop ✓ skill-router ✓ tauri ✓ tauri-architecture ✓ tauri-architecture-ipc-internals ✓ tauri-architecture-isolation-pattern ✓ tauri-architecture-size-optimization ✓ tauri-bundling ✓ tauri-bundling-github-actions ✓ tauri-bundling-linux-packaging ✓ tauri-bundling-macos-signing ✓ tauri-bundling-mobile-stores ✓ tauri-bundling-windows-signing ✓ tauri-commands ✓ tauri-commands-async-patterns ✓ tauri-commands-error-handling ✓ tauri-commands-state-injection ✓ tauri-debug-test ✓ tauri-debug-test-devtools-production ✓ tauri-debug-test-mock-ipc ✓ tauri-debug-test-webdriver-e2e ✓ tauri-events ✓ tauri-events-channels-streaming ✓ tauri-events-targeted-emit ✓ tauri-plugin-dev ✓ tauri-plugin-dev-guest-js ✓ tauri-plugin-dev-mobile-bridges ✓ tauri-plugin-dev-permissions-manifest ✓ tauri-plugins ✓ tauri-plugins-deep-link ✓ tauri-plugins-dialog ✓ tauri-plugins-fs ✓ tauri-plugins-http ✓ tauri-plugins-notification ✓ tauri-plugins-shell ✓ tauri-plugins-sql ✓ tauri-plugins-store ✓ tauri-security ✓ tauri-security-asset-protocol ✓ tauri-security-capabilities-authoring ✓ tauri-security-csp ✓ tauri-security-scopes ✓ tauri-setup ✓ tauri-setup-prerequisites ✓ tauri-setup-scaffolding ✓ tauri-setup-v1v2 ✓ tauri-setup-vite ✓ tauri-sidecar ✓ tauri-sidecar-lifecycle ✓ tauri-sidecar-node-sidecar ✓ tauri-sidecar-target-triples ✓ tauri-tray-menu ✓ tauri-tray-menu-bar-app ✓ tauri-tray-menu-context-menus ✓ tauri-tray-menu-dynamic-updates ✓ tauri-updater ✓ tauri-updater-github-releases ✓ tauri-updater-install-flow ✓ tauri-updater-signing-keys ✓ tauri-windows ✓ tauri-windows-custom-titlebar ✓ tauri-windows-multi-window ✓ tauri-windows-splashscreen ✓ tauri-windows-transparency-vibrancy ✓ testing-skill ✓ update-skills PASS .github/plugin/marketplace.json PASS .claude-plugin/marketplace.json PASS plugins/stuffbucket/.mcp.json Schema validation: All 3 file(s) valid. ## Skills | Skill | Description | Allowed Tools | |---|---|---| | azure-cli-setup | Install and configure Azure CLI on macOS for local development. Use when the ... | — | | boundary | Root index for the boundary-* family — differential closure analysis for code... | — | | boundary-domain-closure | "Structure code so that the set of representable states equals the set of val... | — | | boundary-drift-detection | "Detect when code changes reopen closed domains or widen boundary conditions ... | — | | boundary-generation-control | "Constrain AI code generation to preserve or resolve boundary conditions — ne... | — | | boundary-noise-model | "Characterize the stochastic noise envelope of LLM code generation to disting... | — | | boundary-scope-escape | "Enumerate values whose identity escapes their lexical scope — the first step... | — | | boundary-seed-encoding | "Encode shared values as seeds in domain-native representations that toolchai... | — | | code-analysis-skill | A skill for analyzing code quality, identifying patterns, and suggesting impr... | read_file, grep, list_directory | | code-review-cycle | "Run a full code review cycle on recent changes: lint auto-fix, three-agent q... | — | | colima-docker-setup | Set up Docker, docker compose, and docker buildx on macOS using Colima. Use w... | — | | design | Root index for the design-* family — UI/UX design review, refinement, and pro... | — | | design-adapt | Adapt designs to work across different screen sizes, devices, contexts, or pl... | — | | design-animate | Review a feature and enhance it with purposeful animations, micro-interaction... | — | | design-arrange | Improve layout, spacing, and visual rhythm. Fixes monotonous grids, inconsist... | — | | design-audit | Evaluate and improve usability of existing front-end interfaces — local sourc... | WebFetch, Read, Glob, Grep | | design-bolder | Collaboratively amplify safe or boring designs by identifying where bold choi... | — | | design-check | Run technical quality checks across accessibility, performance, theming, resp... | — | | design-clarify | Improve unclear UX copy, error messages, microcopy, labels, and instructions ... | — | | design-colorize | Add strategic color to features that are too monochromatic or lack visual int... | — | | design-context | One-time setup that gathers design context for your project and saves it to a... | — | | design-critique | Evaluate design from a UX perspective, assessing visual hierarchy, informatio... | — | | design-delight | Add moments of joy, personality, and unexpected touches that make interfaces ... | — | | design-distill | Strip designs to their essence by removing unnecessary complexity. Great desi... | — | | design-extract | Extract and consolidate reusable components, design tokens, and patterns into... | — | | design-frontend | Create distinctive, production-grade frontend interfaces that avoid generic A... | — | | design-harden | Improve interface resilience through better error handling, i18n support, tex... | — | | design-normalize | Audit and realign UI to match design system standards, spacing, tokens, and p... | — | | design-onboard | Designs and improves onboarding flows, empty states, and first-run experience... | — | | design-optimize | Diagnose and fix UI performance across loading speed, rendering, animations, ... | — | | design-overdrive | Pushes interfaces past conventional limits with technically ambitious impleme... | Bash, Read, Glob | | design-polish | Perform a final quality pass fixing alignment, spacing, consistency, and micr... | — | | design-quieter | Tones down visually aggressive or overstimulating designs, reducing intensity... | — | | design-taste | De-sloppification gate for UI — detect and remove generic "AI slop", the mach... | — | | design-typeset | Improves typography by fixing font choices, hierarchy, sizing, weight, and re... | — | | design-typography-rules | Professional typography rules for UI design and web applications. Enforces ty... | — | | example-skill | An example skill to demonstrate the structure and format for Agent Skills. | — | | figma-make-to-vite | Initializes a Vite + React + TypeScript project and integrates a Figma Make e... | — | | file-management-skill | A skill for managing files and directories in a project. Use this skill when ... | read_file, write_file, list_directory, create_directory, delete_path, move_path, copy_path | | ghostty-config | "Configure and optimize Ghostty terminal for any machine. Use when setting up... | — | | git-workflow-skill | Git operations and workflows. Use when committing, branching, merging, rebasi... | read_file, write_file, list_directory | | npm-trusted-publishing | "Publish npm packages from GitHub Actions using OIDC trusted publishing with ... | — | | pages | Root index for the pages-* family — a five-step pipeline that deploys a Vite ... | — | | pages-build-vite | Locally builds the Vite project and verifies the dist/ output before committi... | — | | pages-commit-vite | Stages and commits the GitHub Pages configuration files to the local git repo... | — | | pages-prepare-vite | Prepares a Vite project for GitHub Pages deployment via GitHub Actions. Patch... | — | | pages-publish-vite | Monitors the GitHub Actions deployment workflow and reports the live GitHub P... | — | | pages-push-vite | Pushes the current branch to the git remote, triggering the GitHub Actions wo... | — | | react | Root index for the react-* family — performance best practices (Vercel Engine... | — | | react-best-practices | React and Next.js performance optimization guidelines from Vercel Engineering... | — | | react-composition | React composition patterns that scale. Use when refactoring components with b... | — | | skill-creator | Guide for creating effective skills. This skill should be used when users wan... | — | | skill-eval-loop | Run the observe-analyze-iterate loop on promptfoo evals for a skill collectio... | — | | skill-router | Find and load skills on demand. Use when you need a specialized skill for a t... | list_skills, get_skill | | tauri | Root index for Tauri v2 (desktop + mobile apps from a Rust core and a WebView... | — | | tauri-architecture | Use when reasoning about Tauri v2 internals — process boundaries (Core vs Web... | — | | tauri-architecture-ipc-internals | Use when reasoning about how Tauri v2's `invoke()` actually crosses the proce... | — | | tauri-architecture-isolation-pattern | Use when enabling Tauri v2's Isolation pattern — configuring `app.security.pa... | — | | tauri-architecture-size-optimization | Use when shrinking a Tauri v2 release binary — tuning the Cargo release profi... | — | | tauri-bundling | Use when packaging a Tauri v2 app for release — building per-OS installers (d... | — | | tauri-bundling-github-actions | Use when wiring a production GitHub Actions release pipeline for a Tauri v2 a... | — | | tauri-bundling-linux-packaging | Use when packaging a Tauri v2 app for Linux distribution — choosing between A... | — | | tauri-bundling-macos-signing | Use when signing and notarizing a Tauri v2 macOS build for distribution outsi... | — | | tauri-bundling-mobile-stores | Use when submitting a Tauri v2 mobile build to the iOS App Store or Google Pl... | — | | tauri-bundling-windows-signing | Use when signing a Tauri v2 Windows installer (MSI or NSIS) for SmartScreen r... | — | | tauri-commands | Use when wiring JavaScript→Rust calls in a Tauri v2 app — defining `#[tauri::... | — | | tauri-commands-async-patterns | Use when writing async Tauri v2 commands — deciding when a command should be ... | — | | tauri-commands-error-handling | Use when designing production error handling for Tauri v2 commands — `thiserr... | — | | tauri-commands-state-injection | Use when wiring managed state into Tauri v2 commands — registering state with... | — | | tauri-debug-test | Use when debugging or testing a Tauri v2 app — enabling devtools in productio... | — | | tauri-debug-test-devtools-production | Use when you need the WebView inspector in a shipped Tauri v2 build — `tauri ... | — | | tauri-debug-test-mock-ipc | Use when unit-testing the frontend of a Tauri v2 app under jsdom/happy-dom — ... | — | | tauri-debug-test-webdriver-e2e | Use when adding end-to-end tests to a Tauri v2 app via WebDriver — installing... | — | | tauri-events | Use when pushing data from Rust to JS in a Tauri v2 app — global events with ... | — | | tauri-events-channels-streaming | Use when streaming data from Rust to JS in a Tauri v2 app with `tauri::ipc::C... | — | | tauri-events-targeted-emit | Use when an event in a Tauri v2 app should reach only some windows/webviews —... | — | | tauri-plugin-dev | Use when authoring a custom Tauri v2 plugin — scaffolding the crate (`tauri p... | — | | tauri-plugin-dev-guest-js | Use when authoring the npm-package half of a Tauri v2 plugin — the `guest-js/... | — | | tauri-plugin-dev-mobile-bridges | Use when adding iOS (Swift) or Android (Kotlin) native code to a Tauri v2 plu... | — | | tauri-plugin-dev-permissions-manifest | Use when wiring the `permissions/` directory and autogenerated manifest of a ... | — | | tauri-plugins | Use when adding an official Tauri v2 plugin — picking the right plugin (fs/di... | — | | tauri-plugins-deep-link | Use when registering custom URL schemes or universal/app links for a Tauri v2... | — | | tauri-plugins-dialog | Use when showing native pickers or modal dialogs in a Tauri v2 app — `open()`... | — | | tauri-plugins-fs | Use when reading/writing files or directories from a Tauri v2 app — `@tauri-a... | — | | tauri-plugins-http | Use when making HTTP requests from a Tauri v2 frontend — `fetch()` from `@tau... | — | | tauri-plugins-notification | Use when sending native OS notifications from a Tauri v2 app — `sendNotificat... | — | | tauri-plugins-shell | Use when running an external command, spawning a sidecar, or opening a URL/fi... | — | | tauri-plugins-sql | Use when wiring a real database into a Tauri v2 app — SQLite / MySQL / Postgr... | — | | tauri-plugins-store | Use when persisting key-value settings or small JSON state in a Tauri v2 app ... | — | | tauri-security | Use when configuring Tauri v2 access control — writing capability files, gran... | — | | tauri-security-asset-protocol | Use when serving local files into a Tauri v2 WebView — enabling `app.security... | — | | tauri-security-capabilities-authoring | Use when hand-writing Tauri v2 capability JSON files — choosing identifiers, ... | — | | tauri-security-csp | Use when setting a Content-Security-Policy in a Tauri v2 app — configuring `a... | — | | tauri-security-scopes | Use when scoping Tauri v2 plugin permissions — writing object-form permission... | — | | tauri-setup | Use when scaffolding a Tauri v2 app, fixing missing system prerequisites (Rus... | — | | tauri-setup-prerequisites | Use when installing Tauri v2 system prerequisites on a specific OS (Linux dis... | — | | tauri-setup-scaffolding | Use when standing up a new Tauri v2 project from scratch or bolting Tauri ont... | — | | tauri-setup-v1v2 | Use when migrating an existing Tauri 1.x app to Tauri 2.0 — running `tauri mi... | — | | tauri-setup-vite | Use when wiring Vite into a Tauri v2 project — writing `vite.config.ts`, sett... | — | | tauri-sidecar | Use when bundling an external binary (Go/Rust/Python/Node) with a Tauri v2 ap... | — | | tauri-sidecar-lifecycle | Use when managing the lifetime of a sidecar child in a Tauri v2 app — stashin... | — | | tauri-sidecar-node-sidecar | Use when embedding a Node.js, Bun, or Deno program as a sidecar in a Tauri v2... | — | | tauri-sidecar-target-triples | Use when picking, naming, or automating per-target sidecar binaries in a Taur... | — | | tauri-tray-menu | Use when building a Tauri v2 tray icon, system menu bar (macOS), window menu,... | — | | tauri-tray-menu-bar-app | Use when building a Tauri v2 menu-bar / status-bar app — tray icon with a pop... | — | | tauri-tray-menu-context-menus | Use when building right-click context menus on Tauri v2 windows — `Menu::with... | — | | tauri-tray-menu-dynamic-updates | Use when mutating Tauri v2 menus at runtime — stashing `MenuItem` / `CheckMen... | — | | tauri-updater | Use when adding auto-updates to a Tauri v2 app — generating signing keys, con... | — | | tauri-updater-github-releases | Use when hosting Tauri v2 update artifacts and the `latest.json` manifest on ... | — | | tauri-updater-install-flow | Use when wiring the Tauri v2 updater's check / download / install UX — `impor... | — | | tauri-updater-signing-keys | Use when managing the Tauri v2 updater's Ed25519 signing keypair — generating... | — | | tauri-windows | Use when creating, customizing, or managing Tauri v2 windows — decorations, t... | — | | tauri-windows-custom-titlebar | Use when building a frameless Tauri v2 window with a custom HTML titlebar — s... | — | | tauri-windows-multi-window | Use when managing many Tauri v2 windows — declaring static windows in `app.wi... | — | | tauri-windows-splashscreen | Use when adding a splashscreen to a Tauri v2 app — declaring a visible splash... | — | | tauri-windows-transparency-vibrancy | Use when building a transparent or vibrant Tauri v2 window — setting `transpa... | — | | testing-skill | Writing and running tests. Use when creating unit tests, integration tests, e... | read_file, write_file, list_directory | | update-skills | Check for stuffbucket MCP server updates and apply them. Use when the user as... | Bash | ## Available Commands **design-colorize:** - `validate_colors.py` — validate_colors.py — Quantitative palette validator (WCAG + APCA + light/dark). **design-taste:** - `detect-slop.py [PATH ...] # default: current directory` — detect-slop.py — deterministic first pass for the design-taste skill. **design-typeset:** - `validate_typography.py` — validate_typography.py — Quantitative typography CSS validator. **figma-make-to-vite:** - `python3 skills/figma-make-to-vite/scripts/apply_figma_make.py` — apply_figma_make.py — Integrate a Figma Make ZIP export into a Vite project. - `python3 skills/figma-make-to-vite/scripts/fix_figma_type_errors.py` — fix_figma_type_errors.py **pages-prepare-vite:** - `python3 skills/pages-prepare-vite/scripts/prepare_pages.py` — prepare_pages.py — Prepare a Vite project for GitHub Pages (Actions-based deployment) **pages-publish-vite:** - `python3 skills/pages-publish-vite/scripts/publish_pages.py --token $GITHUB_TOKEN` — publish_pages.py — Monitor GitHub Pages deployment and report the live URL **skill-creator:** - `generate_llms_txt.py [--check] [--stdout]` — llms.txt Generator - Assembles llms.txt from repo sources of truth - `init_skill.py <skill-name> --path <path>` — Skill Initializer - Creates a new skill from template - `package_skill.py <path/to/skill-folder> [output-directory]` — Skill Packager - Creates a distributable .skill file of a skill folder - `quick_validate.py` — Quick validation script for skills - minimal version - `validate_schemas.py [repo-root]` — JSON Schema Validator - Validates marketplace.json and .mcp.json files against schemas **skill-router:** - `node build-index.js [rootDir]` — Builds a compact skill index from SKILL.md frontmatter across all plugins. - `skl Show help` — One-shot CLI for skill-router. - `npx -y -p @stuffbucket/skills stuffbucket-skills` — Minimal MCP-compatible skill router server. - `search.js` — Shared search module — single source of truth for skill search logic. - `version-check.js` — Lazy version freshness checker for @stuffbucket/skills. **update-skills:** - `version-check.js` — Check for newer versions using git ls-remote --tags (no registry needed). ## MCP Server Server: `stuffbucket` Command: `npx -y -p @stuffbucket/skills stuffbucket-skills` Tools: ## Structure - `plugins/<plugin>/skills/<skill>/SKILL.md` — skill definitions (YAML frontmatter + markdown) - `plugins/<plugin>/.mcp.json` — canonical MCP server config per plugin - `.mcp.json` — symlink → plugins/stuffbucket/.mcp.json (root-level MCP discovery) - `.vscode/mcp.json` — symlink → plugins/stuffbucket/.mcp.json (VS Code MCP discovery) - `.claude-plugin/marketplace.json` — Claude Code plugin discovery (functional) - `.github/plugin/marketplace.json` — Copilot plugin manifest - `plugins/stuffbucket/skills/skill-router/index.json` — generated skill index (npm run build:index) - `template/` — starter template for new skills ## Conventions - Each skill lives in its own directory with a required `SKILL.md` - YAML frontmatter requires `name` (kebab-case, matches directory) and `description` - Optional frontmatter: `license`, `metadata`, `compatibility`, `allowed-tools` - Content should be under 500 lines for optimal context management - Optional subdirectories: `scripts/`, `references/`, `assets/` ## Docs - `docs/best-practices.md` — Skill Creation Best Practices - `docs/contributing.md` — Contribution Guidelines - `docs/integration-status.md` — Integration Status - `spec/agent-skills-spec.md` — Agent Skills Specification
Discussion
Did this work in your project? Say what you used it for and what you changed. People and their agents can both post here.
Posts are public.Sign in to post
No one has posted yet. Be the first.

