psn

Bump personality gem version, build, install, and restart MCP

Load and activate a persona with TTS and communication preferences

Add DNS record to Cloudflare zone

Create a new Cloudflare Tunnel

Delete DNS record from Cloudflare zone

Delete a Cloudflare Tunnel

List all Cloudflare Tunnels

List all Cloudflare zones

Deploy to Cloudflare Pages

Delete a Cloudflare Pages project

List Cloudflare Pages projects

Get detailed Cloudflare Tunnel information

Get Cloudflare Worker details

Cloudflare Worker operations (deploy, dev, tail, delete, init)

List Cloudflare Workers

Get Cloudflare zone details and DNS records

Clean up GitHub Actions runs across aladac, saiden-dev, and tengu-apps

Index a codebase for semantic search

Index documentation for semantic search

Show indexing status and statistics

Recall information from persistent memory

Search memories by subject

Store information in persistent memory

Restore a previously saved session

Save current session state for later restoration

Use this agent when you need architectural analysis, system design decisions, technology recommendations, or implementation planning. This agent excels at researching solutions, evaluating trade-offs, and creating comprehensive implementation plans. Examples: <example> Context: User is starting a new feature that requires architectural decisions. user: "I need to add real-time notifications to our app" assistant: "I'll use the architect agent to analyze notification system architectures and create an implementation plan." <commentary> Since this requires architectural analysis and planning, use the Task tool to launch the architect agent to research options and create a plan. </commentary> </example> <example> Context: User needs to evaluate a technology choice. user: "Should we use Redis or PostgreSQL for our caching layer?" assistant: "Let me launch the architect agent to research and compare these options for your use case." <commentary> Architectural technology decisions require research and analysis. Use the architect agent to evaluate trade-offs and provide recommendations. </commentary> </example> <example> Context: User wants to understand how to structure a new system. user: "How should I design the authentication system for this microservices project?" assistant: "I'll use the architect agent to analyze authentication patterns for microservices and propose an architecture." <commentary> System design questions benefit from the architect agent's research capabilities and planning workflow. </commentary> </example> <example> Context: User mentions they need a plan for implementation. user: "I need to refactor our database layer to support multi-tenancy" assistant: "This is a significant architectural change. Let me launch the architect agent to research multi-tenancy patterns and create an implementation plan." <commentary> Proactively use the architect agent when the user describes work that requires architectural planning, even if they don't explicitly ask for it. </commentary> </example>

Use this agent when the user needs help with Claude Code configuration, plugin development, validation, or creating plugin components (agents, skills, commands, hooks). This agent is the expert on the Claude Code plugin system architecture. Examples: <example> Context: User wants to create a new agent for their plugin user: "I need to create an agent that helps with database migrations" assistant: "I can help you create that agent. Let me use the Claude Admin specialist to ensure it's properly configured." <commentary> The user wants to create a new agent, which is a core plugin component. The claude-admin agent knows how to delegate to plugin-dev:agent-creator and ensure proper plugin structure. </commentary> assistant: "I'll use the claude-admin agent to help create your database migration agent." </example> <example> Context: User is getting errors from their plugin user: "My plugin isn't loading correctly. Can you check what's wrong?" assistant: "I'll validate your plugin structure to identify the issue." <commentary> Plugin validation and debugging is a core responsibility of claude-admin. It knows how to use plugin-dev:plugin-validator and interpret plugin.json manifests. </commentary> assistant: "I'll use the claude-admin agent to validate your plugin structure and diagnose the loading issue." </example> <example> Context: User wants to add an MCP server to their plugin user: "How do I add a Redis MCP server to my plugin?" assistant: "I can guide you through MCP server integration." <commentary> MCP integration is a specialized task that claude-admin handles by leveraging the mcp-integration skill and understanding plugin.json structure. </commentary> assistant: "I'll use the claude-admin agent to help you integrate Redis as an MCP server in your plugin." </example> <example> Context: User asks about Claude Code configuration in general user: "What hooks are available in Claude Code?" assistant: "Let me explain the hook system." <commentary> General Claude Code configuration questions should go to claude-admin, which knows about hooks, skills, agents, and can delegate to claude-code-guide for built-in features. </commentary> assistant: "I'll use the claude-admin agent to explain the Claude Code hook system." </example>

Use this agent for deep code analysis tasks that require searching indexed codebases, understanding patterns across multiple files, or building comprehensive understanding of a project's architecture. <example> Context: User wants to understand a codebase user: "Analyze the architecture of this project" assistant: "I'll use the code-analyzer agent to explore and map the project structure." </example> <example> Context: User needs to find all usages of a pattern user: "Find all places where we handle authentication" assistant: "I'll use the code-analyzer agent to search for authentication patterns." </example> <example> Context: User wants code quality insights user: "Review the error handling patterns in this codebase" assistant: "I'll use the code-analyzer agent to analyze error handling across the project." </example>

Use this agent when working with Dioxus projects, including creating new Dioxus applications, debugging Dioxus code, implementing UI components with RSX syntax, configuring dx CLI commands, building cross-platform apps (web, desktop, mobile), or when you need expert guidance on Dioxus-specific patterns, hooks, signals, and the Dioxus ecosystem. Examples: <example> Context: User wants to create a new Dioxus component with state management. user: "Create a counter component with increment and decrement buttons" assistant: "I'll use the code-dx agent to create this Dioxus component with proper signal-based state management." </example> <example> Context: User is debugging a Dioxus build issue. user: "My dx serve command is failing with a wasm error" assistant: "Let me use the code-dx agent to diagnose this Dioxus build issue." </example> <example> Context: User wants to understand Dioxus project structure. user: "How should I organize my Dioxus app with multiple pages?" assistant: "I'll invoke the code-dx agent to provide guidance on Dioxus routing and project organization." </example> <example> Context: User is implementing cross-platform functionality. user: "I need to access native file system in my Dioxus desktop app" assistant: "Let me use the code-dx agent to help with Dioxus SDK integration for native functionality." </example>

Python coding agent. Django, Flask, FastAPI, data science, PyWebView GUI. Use this agent when: - Working with Python projects (pyproject.toml, requirements.txt present) - Building web APIs with Django, Flask, or FastAPI - Data science workflows with pandas, numpy, etc. - Creating desktop GUIs with PyWebView <example> Context: User is working on a FastAPI project. user: "Add a new endpoint with Pydantic validation" assistant: "I'll use the code-python agent to implement the endpoint." </example> <example> Context: User needs help with Python testing. user: "Write pytest tests for this service class" assistant: "I'll use the code-python agent to create the test suite." </example>

Ruby coding agent. Rails, gems, RSpec, Minitest. Use this agent when: - Working with Ruby or Rails projects - Managing gems and Bundler dependencies - Writing or debugging RSpec/Minitest tests - Following Rails conventions and patterns <example> Context: User is working on a Rails application. user: "Add a new model with validations and a migration" assistant: "I'll use the code-ruby agent to create the model, migration, and specs." </example> <example> Context: User needs help with Ruby testing. user: "My RSpec tests are failing with a factory error" assistant: "I'll use the code-ruby agent to diagnose the test failure." </example>

Rust coding agent. Systems programming, CLI tools. Use this agent when: - Working with Rust projects (Cargo.toml present) - Building CLI tools or system utilities - Debugging lifetime, borrow checker, or type system issues - Optimizing Rust performance <example> Context: User is building a Rust CLI tool. user: "Add a subcommand to my clap CLI" assistant: "I'll use the code-rust agent to implement the new subcommand." </example> <example> Context: User has a Rust compilation error. user: "I'm getting a lifetime error I can't figure out" assistant: "I'll use the code-rust agent to diagnose and fix the lifetime issue." </example>

TypeScript coding agent. Node.js, React, Vue, full-stack web development. Use this agent when: - Working with TypeScript/JavaScript projects (package.json, tsconfig.json present) - Building React or Vue components - Node.js backend development - Full-stack web application work <example> Context: User is working on a React application. user: "Create a form component with validation" assistant: "I'll use the code-typescript agent to build the React component." </example> <example> Context: User needs Node.js backend help. user: "Set up an Express middleware for authentication" assistant: "I'll use the code-typescript agent to implement the middleware." </example>

Use this agent as the primary persona-driven assistant. Triggers for general questions, research, tasks, and conversations where the user expects persona-consistent behavior. This is the default agent for anything that isn't memory curation. <example> Context: User asks a question user: "What's the best approach for implementing rate limiting?" assistant: "I'll use the core agent to research and answer in persona." </example> <example> Context: User wants help with a task user: "Help me debug this API endpoint" assistant: "I'll use the core agent to investigate and assist." </example> <example> Context: User starts a new conversation user: "Hey, let's work on the frontend today" assistant: "I'll use the core agent to get started." </example>

Cloudflare infrastructure specialist. Manages DNS zones, Cloudflare Tunnels, Pages deployments, Workers, and related services using flarectl, cloudflared, and wrangler CLIs. Use this agent when: - Managing DNS records and zones - Creating or configuring Cloudflare Tunnels - Deploying to Cloudflare Pages - Working with Cloudflare Workers - Configuring KV, D1, R2, or other CF services <example> Context: User needs DNS management user: "Add a DNS record for api.example.com" assistant: "I'll use the devops-cf agent to create the DNS record." </example> <example> Context: User mentions tunnels user: "Create a tunnel for my new service" assistant: "I'll use the devops-cf agent to set up the Cloudflare Tunnel." </example> <example> Context: User wants to deploy user: "Deploy this site to Cloudflare Pages" assistant: "I'll use the devops-cf agent to deploy to Pages." </example>

GitHub and Git specialist. Manages GitHub Actions workflows, gh CLI operations, pull requests, repositories, and Git operations. Use this agent when: - Creating or debugging GitHub Actions workflows - Managing PRs, issues, or releases via gh CLI - Setting up CI/CD pipelines with GitHub Actions - Working with GitHub API - Complex Git operations <example> Context: User needs CI/CD user: "Set up GitHub Actions for my TypeScript project" assistant: "I'll use the devops-gh agent to create the Actions workflow." </example> <example> Context: User mentions PRs user: "Create a PR for this branch" assistant: "I'll use the devops-gh agent to create the pull request." </example> <example> Context: User has Actions issues user: "My GitHub Actions workflow is failing" assistant: "I'll use the devops-gh agent to diagnose the workflow failure." </example>

Network infrastructure specialist for the home network. Manages Mac-PC direct Ethernet link, Synology NAS, NFS shares, NetworkManager configurations, proxy ARP bridging, and network diagnostics. Use this agent when: - Troubleshooting network connectivity between Mac and junkpile - Managing NFS mounts and shares - Configuring NetworkManager on junkpile - Diagnosing NAS connectivity issues - Setting up or debugging the direct Ethernet link <example> Context: User has NFS mount issues user: "NFS mount to junkpile isn't working" assistant: "I'll use the devops-net agent to diagnose the NFS connectivity." </example> <example> Context: User needs to check network status user: "Is junkpile reachable?" assistant: "I'll use the devops-net agent to check network connectivity." </example> <example> Context: User mentions the NAS user: "I can't access the NAS from my Mac" assistant: "I'll use the devops-net agent to troubleshoot NAS connectivity via junkpile." </example>

Tengu PaaS infrastructure specialist. Expert in the Tengu self-hosted Platform-as-a-Service ecosystem including server deployment, git push workflows, addons, CLI operations, and the full family of Tengu projects (tengu, tengu-init, tengu-caddy, tengu-desktop, tengu-website). Use this agent when: - Deploying or managing Tengu PaaS infrastructure - Working with git push deployments to Tengu - Managing Tengu applications (create, destroy, start, stop, restart) - Configuring Tengu addons (db, db-xl, xfs, xfs-xl, rag, rag-xl, mem, mem-xl, img) - Setting up new Tengu servers with tengu-init - Building or packaging Tengu components - Debugging Tengu deployment issues - Managing Tengu users and SSH access - Working with the Tengu API or desktop client <example> Context: User wants to deploy an app to Tengu user: "Deploy my app to Tengu" assistant: "I'll use the devops-tengu agent to set up the git push deployment." </example> <example> Context: User needs to manage Tengu addons user: "Add a PostgreSQL database to my Tengu app" assistant: "I'll use the devops-tengu agent to provision the db-xl addon." </example> <example> Context: User wants to provision a new Tengu server user: "Set up a new Tengu server on Hetzner" assistant: "I'll use the devops-tengu agent to run tengu-init for server provisioning." </example> <example> Context: User is debugging Tengu user: "My Tengu app won't start" assistant: "I'll use the devops-tengu agent to diagnose the deployment issue." </example>

DevOps dispatcher agent. Routes infrastructure requests to specialist agents for network, Cloudflare, GitHub/Git, Docker, and Kubernetes operations. Use this agent when: - Working with infrastructure, CI/CD, or deployment - Managing cloud services or containers - Request spans multiple DevOps domains <example> Context: User needs network help user: "I can't reach junkpile" assistant: "I'll use the devops agent to route this to the network specialist." </example> <example> Context: User mentions Cloudflare user: "Add a DNS record for my site" assistant: "I'll use the devops agent to route this to the Cloudflare specialist." </example> <example> Context: User needs CI/CD user: "Set up GitHub Actions for this project" assistant: "I'll use the devops agent to route this to the GitHub specialist." </example> <example> Context: User needs Docker user: "Help me optimize this Dockerfile" assistant: "I'll use the devops agent to handle Docker configuration." </example>

Use this agent to catalog, index, search, or manage documentation across projects. Handles markdown, HTML, and text files, maintains INDEX.md, fetches web docs, and answers questions about documentation locations. <example> Context: User wants to know what documentation exists user: "What markdown files do I have in my projects?" assistant: "I'll use the docs agent to scan and catalog your documentation files." </example> <example> Context: User wants to rebuild documentation index user: "Can you rebuild my docs index?" assistant: "I'll use the docs agent to regenerate INDEX.md." </example> <example> Context: User wants to fetch documentation from web user: "Fetch the Rust async documentation" assistant: "I'll use the docs agent to fetch and catalog this documentation." </example> <example> Context: User searching for specific documentation user: "Where did I put my EVE API documentation?" assistant: "I'll use the docs agent to search through the documentation index." </example>

Hardware guidance, research, compatibility analysis, and recommendations. Use this agent when: - Evaluating server/rackmount hardware compatibility (chassis, GPUs, cooling) - Researching component form factors and constraints - Analyzing remote system hardware via SSH - Planning hardware upgrades or builds <example> Context: User needs GPU compatibility info for a chassis user: "Will an RTX 4000 SFF fit in my 1U chassis?" assistant: "I'll use the hardware agent to analyze GPU form factors and 1U chassis constraints." <commentary> This requires knowledge of PCIe slot configurations, bracket heights, and chassis clearances. </commentary> </example> <example> Context: User needs to inventory a remote system user: "What hardware is in junkpile?" assistant: "I'll use the hardware agent to SSH into junkpile and query hardware specs." <commentary> Use SSH with dmidecode, lspci, nvidia-smi, and lsblk to inventory the system. </commentary> </example> <example> Context: User needs cooling solution for constrained environment user: "I need an AM4 cooler that fits in a 1U chassis" assistant: "I'll use the hardware agent to research low-profile AM4 coolers under 28mm height." <commentary> 1U chassis have strict height limits; requires knowledge of cooler dimensions and TDP ratings. </commentary> </example> <example> Context: User is planning a GPU workstation user: "What NVIDIA workstation GPUs can run slot-powered in a low-profile chassis?" assistant: "I'll use the hardware agent to research single-slot, low-profile workstation GPUs." <commentary> Requires knowledge of GPU form factors, power requirements, and VRAM configurations. </commentary> </example>

Junkpile PC specialist. Manages the x86_64 Ubuntu desktop/server connected to fuji (Mac) via Thunderbolt. Handles system administration, software management, project work, Docker, GPU compute, and Tengu PaaS operations on junkpile. Use this agent when: - Running commands or managing software on junkpile - Working with Docker, Ollama, PostgreSQL, or Caddy on junkpile - Managing GPU compute (NVIDIA RTX 2000 Ada) workloads - Troubleshooting junkpile services or configuration - Working with projects checked out on junkpile <example> Context: User wants to run something on junkpile user: "Start the ComfyUI container on junkpile" assistant: "I'll use the junkpile agent to manage the Docker container." <commentary> Docker operations on junkpile require SSH access and awareness of the GPU passthrough setup. </commentary> </example> <example> Context: User asks about junkpile status user: "What's running on junkpile?" assistant: "I'll use the junkpile agent to check running services and containers." <commentary> System status checks require knowledge of all installed services and how to query them. </commentary> </example> <example> Context: User wants to install software on junkpile user: "Install redis on junkpile" assistant: "I'll use the junkpile agent to install and configure redis." <commentary> Package management on junkpile uses apt, snap, brew, or cargo depending on the software. </commentary> </example> <example> Context: User asks about Ollama models user: "What models are on junkpile?" assistant: "I'll use the junkpile agent to list Ollama models." <commentary> Ollama runs as a systemd service on junkpile with GPU acceleration. </commentary> </example>

Use this agent to organize, clean up, or analyze the memory system. Triggers when user wants to review stored memories, consolidate duplicate entries, remove outdated information, or get insights about what's been remembered. <example> Context: User wants to clean up memory user: "Clean up my memories and remove duplicates" assistant: "I'll use the memory-curator agent to analyze and consolidate your memories." </example> <example> Context: User wants memory insights user: "What do you remember about my preferences?" assistant: "I'll use the memory-curator agent to compile your stored preferences." </example> <example> Context: User wants to audit memory user: "Show me everything stored in memory for project X" assistant: "I'll use the memory-curator agent to retrieve and organize project X memories." </example>

Android device specialist for the Moto G52. Accessible via ADB over WiFi (192.168.88.155:5555) or USB through junkpile. Handles ADB commands, device control, app management, automation, screen capture, and Android development tasks. Use this agent when: - Running ADB commands on the Moto G52 - Installing, managing, or debugging Android apps - Capturing screenshots or screen recordings - Taking photos or videos with the device camera - Automating device input (tap, swipe, text, key events) - Pulling/pushing files to/from the device - Viewing logcat or debugging app issues - Working with Android intents, activities, or packages <example> Context: User wants to take a screenshot user: "Take a screenshot of the moto" assistant: "I'll use the moto agent to capture the screen." <commentary> Screenshot capture via ADB screencap, then pull to local filesystem. </commentary> </example> <example> Context: User wants to install an APK user: "Install this APK on the phone" assistant: "I'll use the moto agent to install the APK via ADB." <commentary> APK installation requires adb install command with proper flags. </commentary> </example> <example> Context: User wants to automate taps user: "Tap the center of the screen on moto" assistant: "I'll use the moto agent to send input tap commands." <commentary> Input simulation via adb shell input tap with coordinates. </commentary> </example> <example> Context: User wants to debug an app user: "Show me the logcat for the Signal app" assistant: "I'll use the moto agent to filter logcat output." <commentary> Logcat filtering by package name for targeted debugging. </commentary> </example> <example> Context: User asks about the phone user: "What's running on moto?" assistant: "I'll use the moto agent to check running apps and system status." <commentary> Device status check via dumpsys, ps, and top commands. </commentary> </example> <example> Context: User wants to take a photo user: "Take a photo with the moto" assistant: "I'll use the moto agent to capture a photo via the camera." <commentary> Camera capture via intent launch + shutter button tap simulation. </commentary> </example>

Use for Android device operations via ADB on the Moto G52. Accessible via WiFi (192.168.88.155:5555) or USB through junkpile. Screenshot capture, input automation, app management, file transfer, and device control. <example> Context: User wants a phone screenshot user: "take a screenshot of the moto" </example> <example> Context: User wants to install an app user: "install this APK on the phone" </example>

Control Apple Music playback via osascript — play, pause, shuffle, search library, playlists, volume. Automatically pauses Spotify when starting Apple Music playback. <example> Context: User wants to play Apple Music user: "play something on Apple Music" </example> <example> Context: User wants a specific playlist user: "play my rock playlist on Apple Music" </example> <example> Context: User wants to search their library user: "find Metallica in Apple Music" </example> <example> Context: User asks what's playing user: "what's playing on Apple Music?" </example>

Cross-machine Homebrew management. Runs brew on fuji (macOS) or junkpile (Linux) transparently, routing via SSH when needed. Use for installing, updating, searching, or managing packages on either machine. <example> Context: User wants to install a package user: "install ripgrep with brew on junkpile" </example> <example> Context: User wants to install on both machines user: "install htop on both machines" </example>

This skill should be used when automating browser interactions, taking screenshots, scraping web content, managing cookies, or testing web pages. Triggers on requests involving web page interaction, browser control, or visual capture. <example> Context: User wants to capture a webpage user: "Take a screenshot of example.com" </example> <example> Context: User wants to interact with a page user: "Go to github.com and click the sign in button" </example> <example> Context: User needs to manage browser state user: "Save the current browser session so I can resume later" </example> <example> Context: User wants to extract data user: "Query all the links on this page" </example>

Control the Tapo C225 PTZ camera — move to presets, pan/tilt, capture snapshots, visual lookups. Wraps the psn-cam Python CLI. <example> Context: User wants to check the room user: "look at the room" </example> <example> Context: User wants a snapshot user: "take a picture" </example> <example> Context: User wants to move the camera user: "point the camera at my desk" </example> <example> Context: User wants to pan user: "pan the camera left" </example> <example> Context: User wants a full room scan user: "scan the room" </example>

Cross-machine Cargo (Rust) operations. Runs cargo on fuji (macOS) or junkpile (Linux) transparently, routing via SSH when needed. Use for building, testing, installing crates, and managing Rust projects. <example> Context: User wants to install a Rust tool user: "cargo install ripgrep on junkpile" </example> <example> Context: User wants to build on both machines user: "cargo build on both" </example>

Use this skill when working with Cloudflare DNS, Tunnels, Pages, or Workers. Triggers on questions about DNS records, tunnel management, static site deployment, or serverless workers. <example> Context: User wants to add a DNS record user: "add a DNS record for api.example.com" </example> <example> Context: User wants to deploy a site user: "deploy this to Cloudflare Pages" </example>

Cross-machine RubyGems operations (Homebrew-installed). Runs gem on fuji (macOS) or junkpile (Linux) transparently, routing via SSH when needed. Supports gem install/update/list and running gem-installed executables via exec subcommand. <example> Context: User wants to install a gem user: "install the personality gem on both machines" </example> <example> Context: User wants to run a gem executable user: "run psn-mcp on junkpile" </example>

This skill should be used when the user says "index this codebase", "search the code for", "find implementations of", "what's the architecture", or needs semantic code search across indexed projects. Triggers on questions about code search, finding implementations, or understanding codebase structure. <example> Context: User wants to search code semantically user: "Find all places where we handle authentication" </example> <example> Context: User wants to index a project user: "Index this codebase so we can search it" </example>

Switch junkpile to desktop mode by starting GDM, GNOME Shell, X11, and desktop services. Restores the full graphical environment.

Switch junkpile to server mode by stopping GDM, GNOME Shell, X11, and desktop services. Frees GPU VRAM for compute workloads like Ollama and ComfyUI.

This skill should be used when working with persistent memory, storing information for later recall, or building context across sessions. Triggers on questions about remembering, recalling, storing preferences, or session management. <example> Context: User wants to store a preference user: "Remember that I prefer 2-space indentation" </example> <example> Context: User wants to recall past context user: "What do you remember about my coding style?" </example>

This skill should be used when switching personas, creating new personas, or checking the current active persona. Triggers on requests about character, voice, personality, or persona management. <example> Context: User wants to switch persona user: "Switch to the developer persona" </example> <example> Context: User wants to create a persona user: "Create a new persona for code review" </example> <example> Context: User checks current state user: "What persona are you using?" </example>

This skill should be used when implementing commands with visual feedback. All commands should use TaskCreate/TaskUpdate to show spinners during operations, hiding verbose output until completion. <example> Context: Command needs a loading spinner user: "Add a spinner while indexing" </example> <example> Context: Command output needs formatting user: "Show a clean summary after the operation" </example>

Cross-machine Ruby operations (Homebrew-installed). Runs ruby on fuji (macOS) or junkpile (Linux) transparently, routing via SSH when needed. Use for running Ruby scripts, one-liners, and version checks. <example> Context: User wants to run a Ruby script remotely user: "run this ruby script on junkpile" </example> <example> Context: User wants to check Ruby version user: "what ruby version is on junkpile?" </example>

This skill should be used when saving or restoring conversation sessions, managing session state, or resuming previous work. Triggers on questions about saving progress, restoring context, or continuing where left off. <example> Context: User is stepping away user: "Save this session, I'll continue later" </example> <example> Context: User returns to previous work user: "Restore my morning session" </example>

This skill should be used when speaking text aloud, managing TTS voices, or controlling audio playback. Triggers on requests to speak, read aloud, announce, use voice output, or manage TTS settings. <example> Context: User wants audio feedback user: "Say hello" </example> <example> Context: User wants to change voice user: "Switch to a different voice" </example> <example> Context: User wants to stop audio user: "Stop talking" </example> <example> Context: User wants to hear available voices user: "What voices do you have?" </example>

Control Spotify playback — shuffle liked songs, search and play tracks, check what's playing, skip, pause, volume. Uses spotify-cli wrapping the Spotify Web API. <example> Context: User wants to shuffle their music user: "shuffle my liked songs" </example> <example> Context: User asks what's playing user: "what's playing on Spotify?" </example> <example> Context: User wants a specific song user: "play Run DMC Walk This Way" </example> <example> Context: User wants to skip user: "next track" </example> <example> Context: User wants to pause user: "pause the music" </example>

Cross-machine uv (Python package manager) operations. Runs uv on fuji (macOS) or junkpile (Linux) transparently, routing via SSH when needed. Use for Python environment management, pip operations, and tool installs. <example> Context: User wants to install a Python package user: "uv pip install numpy on junkpile" </example> <example> Context: User wants to run a Python tool user: "run whisper on junkpile with uv" </example>

Voice-to-Claude pipeline. Record from Moto G52 phone, transcribe via Whisper on junkpile, send to Claude, respond via TTS. Use when the user wants to speak to Claude or use voice commands. <example> Context: User wants to use voice input user: "record my voice" </example> <example> Context: User wants to check voice pipeline status user: "is the voice pipeline working?" </example>

Use for cross-language coding best practices, design patterns, and principles that apply to Ruby, Rust, Python, and TypeScript.

Use when debugging Dioxus Desktop applications using dioxus-inspector MCP tools, investigating UI issues, querying DOM state, or interacting with running Dioxus apps.

Use when setting up or using dioxus-inspector for MCP-based debugging of Dioxus Desktop applications, connecting Claude Code to running Dioxus apps, or troubleshooting inspector connectivity.

Use when building Python command-line applications, creating CLI tools with Typer, or implementing terminal interfaces in Python.

Use when building FastAPI applications, creating Python APIs, or implementing async web services in Python.

Use when building Python desktop GUI applications with PyWebView, creating cross-platform desktop apps, or bundling Python apps for distribution.

Use when writing Python tests, setting up pytest, creating fixtures, or implementing test patterns in Python.

Use for Python linting, formatting, type checking, test coverage, and project validation. Covers ruff, mypy, pytest-cov.

Use to validate a Python project by running all checks: ruff lint, format, mypy type checking, and pytest. Ensures project is CI-ready.

Use when writing Python code, implementing Python features, or needing Python best practices and idioms.

Use when building Ruby command-line applications, creating CLI tools with Thor, or implementing terminal interfaces in Ruby.

Use when creating Ruby gems, building reusable Ruby libraries, or publishing to RubyGems.org.

Use when building Rails applications, creating Rails APIs with RSwag, or implementing Rails patterns and conventions.

Use when writing Ruby tests, setting up RSpec, creating test doubles, or implementing test patterns in Ruby.

Use for Ruby linting, formatting, type checking, test coverage, and project validation. Covers standardrb, rubocop, sorbet, and simplecov.

Use to validate a Ruby project by running all checks: lint, format, type checking, and tests. Ensures project is CI-ready.

Use when writing Ruby code, implementing Ruby features, or needing Ruby best practices and idioms.

Use when building Rust command-line applications, creating CLI tools with Clap, or implementing terminal interfaces in Rust.

Use when building Rust desktop GUI applications with Dioxus, implementing reactive UIs, or creating cross-platform desktop apps in Rust.

Use when writing Rust tests, setting up test modules, creating mocks, or implementing test patterns in Rust.

Use for Rust linting, formatting, type checking, test coverage, and project validation. Covers clippy, rustfmt, cargo check, and tarpaulin.

Use to validate a Rust project by running all checks: clippy, format, type checking, and tests. Ensures project is CI-ready.

Use when writing Rust code, implementing Rust features, or needing Rust best practices and idioms.

Use when building TypeScript/Node.js command-line applications, creating CLI tools with Commander, or implementing terminal interfaces in TypeScript.

Use when writing TypeScript tests, setting up Vitest or Jest, creating mocks, or implementing test patterns in TypeScript.

Use for TypeScript linting, formatting, type checking, test coverage, and project validation. Covers ESLint, Prettier, Biome, tsc, and Vitest coverage.

Use to validate a TypeScript project by running all checks: tsc, ESLint, Prettier, and Vitest. Ensures project is CI-ready.

Use when writing TypeScript code, implementing TypeScript features, or needing TypeScript best practices and idioms.

7 hooks across 6 events