johnlindquist

Always search before starting any work across all coding agent session histories (Claude Code, Codex, Cursor, Gemini CLI, Aider, ChatGPT) to find whatever we've discussed before.

Spawn 5 Opus subagents with randomly-generated distinct personas to debate a problem from multiple angles. Use when exploring UX decisions, architecture choices, or any decision that benefits from diverse perspectives arguing creatively.

Configure Karabiner-Elements keyboard remapping using Goku EDN syntax. Use when creating keybindings, layers, simlayers, app-specific shortcuts, or modifying karabiner.edn.

Bundle code context for AI. ALWAYS use --limit 49k unless user explicitly requests otherwise. Use for creating shareable code bundles and preparing context for LLMs.

Build Raycast extensions with React and TypeScript. Use when the user asks to create a Raycast extension, command, or tool.

Create new Agent Skills for Claude Code. Use when user wants to create a skill, add a new capability, document a CLI workflow, or asks how skills work.

Build durable, resumable TypeScript workflows with Vercel Workflow DevKit. Use when creating long-running processes, AI agents, background jobs, multi-step pipelines, webhooks, or event-driven systems. Triggers on "workflow", "durable", "resumable", "use workflow", "use step".

Track tasks and issues using the bd CLI. Use for task management, sprint planning, dependency tracking, and project organization. Replaces TodoWrite.

Agent-facing DevTools for Script Kit GPUI: use protocol, MCP, and CLI primitives to inspect, control, measure, debug, benchmark, and prove real app UI behavior from bug reports without defaulting to prewritten agentic-testing recipes.

Manage the Notes window — creating, searching, editing, and automating notes via the SDK and automation protocol.

Use the local agy CLI as a fast Script Kit GPUI app inspector by prompting it to drive existing script-kit-devtools primitives, capture logs, and produce a compact investigation result from a user bug report or inspection prompt.

Add Actions Menu commands to scripts and scriptlet bundles with setActions() and companion .actions.md files.

Configure external MCP servers for scripts and Agent Chat. Use when the user wants to add, remove, enable, disable, or inspect MCP servers in Script Kit.

Create or update scriptlet bundles for text expansions, snippets, shell commands, and lightweight helpers in a single markdown file.

Start and manage Agent Chat conversations from scripts, including typed context parts, streaming status, and chat lifecycle operations.

Update Script Kit configuration files — hotkeys, theme, layout, dictation, Claude Code, and workspace settings.

Add purposeful motion to a UXSpec — $animations, onEnter choreography, and timing tokens

Improve every piece of text in a UXSpec — descriptions, labels, error messages, empty states, and loading copy

Identify repeated patterns and consolidate them into $elements and shared tokens for systematic reuse

Collect design context, audience, intent, and requirements before writing any UXSpec JSON

Strengthen a UXSpec against real-world conditions — overflow, missing data, network failures, and accessibility

Run a systematic quality audit on a completed UXSpec — design intent, consistency, accessibility, and state coverage

Walk through building a UXSpec (.uxspec.json) file — gathering requirements, making design decisions about layout/typography/color/motion/accessibility, designing the state machine, defining visuals, then compiling and verifying the output.

Design the $machine section — map happy paths, error paths, and all state types into a complete behavioral skeleton

Development environment setup and workflow for Universe 2025 (Tufty) Badge with MonaOS. Use when setting up the badge, flashing firmware, debugging, or working with the development toolchain.

Create MicroPython applications for Universe 2025 (Tufty) Badge including display graphics, button handling, and MonaOS app structure. Use when building badge apps, creating interactive displays, or developing MicroPython programs.

Deployment workflows, file management, and project organization for Universe 2025 (Tufty) Badge. Use when deploying apps to MonaOS, managing files on device, syncing projects, organizing code, or setting up deployment pipelines.

System diagnostics, verification, and troubleshooting for Badger 2350. Use when checking firmware version, verifying installations, diagnosing hardware issues, troubleshooting errors, or performing system health checks on Badger 2350.

Hardware integration for Badger 2350 including GPIO, I2C sensors, SPI devices, and electronic components. Use when connecting external hardware, working with sensors, controlling LEDs, reading buttons, or interfacing with I2C/SPI devices on Badger 2350.

Complete getting started guide for Universe 2025 (Tufty) Badge from zero to first app. Use when helping absolute beginners, providing step-by-step first-time setup, or when users ask "how do I get started", "where do I begin", or "first steps with the badge".

MicroPython REPL usage, package management, module inspection, and interactive debugging for Universe 2025 (Tufty) Badge. Use when installing MicroPython packages, testing code interactively, checking installed modules, or using the REPL for development.

Python environment setup on your computer for Badger 2350 development. Use when installing Python, setting up virtual environments, installing development tools like mpremote or ampy, or configuring the computer-side development environment for Badger 2350 projects.

Control and inspect Chrome browser using Chrome DevTools Protocol - navigate pages, debug network requests, analyze performance, take screenshots, interact with elements, and automate browser tasks.

Maintain and synchronize Unified Impact Diagrams following Diagram Driven Development (DDD) methodology. Create, update, and manage diagrams that connect user value to technical implementation.

Centralizes persona-driven code reviews (Fowler, Torvalds, Carmack, React core, etc.) so Claude can pick or combine expert viewpoints when the user asks for a code review or perspective-specific critique.

Bundle optimized code context with packx then send it to Oracle's MCP consult tool for GPT-5.4 Pro analysis. Use when the user wants to ask oracle about code with smart filtering, or run "packx oracle", "oracle with packx", or "bundle and ask oracle".

Use packx plus Oracle's MCP consult tool to run iterative GPT-5.4 Pro code reviews in the same ChatGPT browser conversation. Use when the user wants "ask oracle again", "iterate with oracle", "continue the oracle thread", or repeated packx-backed reviews that should preserve prior ChatGPT thread context.

Build and maintain a project's ./flows directory, the agent roster. Each flow is one markdown file that runs as an AI agent (frontmatter + prompt body) on claude, codex, pi, cursor-agent, copilot, agy, or any CLI engine, with colocated behavioral evals. Use when the user asks to "create a flow", "add an agent to this repo", "set up ./flows", "make a markdown agent", or wants repeatable AI tasks with evals.

Creates durable, resumable workflows using Vercel's Workflow DevKit. Use when building workflows that need to survive restarts, pause for external events, retry on failure, or coordinate multi-step operations over time. Triggers on mentions of "workflow", "durable functions", "resumable", "workflow devkit", or step-based orchestration.

Analyze the local Toolsmith corpus of Codex tool calls, identify tool optimization opportunities and blindspots, propose new tools or skills, and suggest focused AGENTS.md/software-engineering improvements based on observed agent behavior.

Publish a skill to GitHub for discovery via `npx skills`. Use when asked to "publish skill", "share skill", "make skill discoverable", or "upload skill to GitHub".

Use when the user asks Codex to consult, run, or reason with the local fusion, fusion-med, or fusion-high CLI for multi-provider feedback, architecture review, planning, critique, or synthesis. Use the local CLI directly, verify providers first, and run the full configured panel to terminal provider states by default, with quota/rate-limit failures treated as non-blocking for that run.

Analyze the local Better Tools corpus of Codex tool calls, identify tool optimization opportunities and blindspots, propose new tools or skills, and suggest focused AGENTS.md/software-engineering improvements based on observed agent behavior.