Explore skills

Encodes and decodes Motorola FLEX paging messages in audio WAV format. FLEX is a high-capacity successor to POCSAG, widely used by commercial paging carriers and active on VHF/UHF networks, supporting various message types.

Captures a secret from the user via a hidden-input dialog and routes it to a specified destination (e.g., 1Password, GitHub secret, .env file) without the value appearing in logs or transcripts. It auto-triggers when the agent requires a new credential, API key, or secret.

Skill that gives AI coding agents persistent memory and quality enforcement

AI experimentation advisor that turns every decision into a testable experiment via Claude Code.

Builds and maintains a navigable graph of iOS, macOS, or Android app screens. It understands navigation flows, instruments screen files, and drives end-to-end flows using console logs and the accessibility tree for testing, without screenshots.

Convenes AI avatars of multiple historical/intellectual figures for a roundtable, engaging in real multi-round discussions on user topics to provide an integrated solution. It uses sub-agents to invoke each character's perspective skill, generating dialogues with substantive confrontation.

Coordinates multiple Claude Code terminals on the same project without collisions. It's ideal for running multiple Claude sessions in parallel, managing multi-agent workflows, and preventing simultaneous file edits, featuring an interactive setup wizard and file-level lock protocol.

A toolkit for cold-starting and promoting open-source GitHub repositories, outlining a four-stage process to take them from completion to popularity, distinguishing AI-driven and manual tasks.

Converts MasterGo designs into frontend component code.

Adds real-time multi-user online commenting to HTML demos. Users can place pins, reply, and change status anywhere on the page, with all changes instantly synchronized via Supabase free tier.

Use this skill when the user requests a real bitmap file (PNG/JPEG) such as an image, illustration, icon, or logo concept. It generates AI raster images via the user's locally-installed Codex CLI (gpt-image-2 model) on their Mac.

Conducts mock interviews and polishes resumes using local files, accumulating interview history and weaknesses across sessions. It offers three modes—interview, polish, debrief—that share resume and transcript state on disk.