Explore AI Agent Skills & Claude Prompts

Voicebox — Open-source AI voice studio for cloning, dictation, and voice creation workflows. Use when the user mentions or needs: Voicebox; open source voice clone studio; local voice AI workstation.

ltx-2

LTX-2 — Official Python inference and LoRA trainer package for the LTX-2 audio-video generative model (Lightricks). Use when the user mentions or needs: LTX-2; Lightricks LTX; LTX-2 inference; LTX-2 LoRA training; audio-video generative model Python.

nephew-cloak-amending-mission

Nephew CLOAK executive layer — full Amending Mission. When Grok or another AI designed systems (whitepapers, agents, repos, merges) but only simulated git, systematically verify every claim against GitHub, rebuild missing artifacts, witness commits, and maintain a green-check ledger. Internalizes Automata as Layer-0 belief system. Triggers: Amending Mission, simulated pushes, fix Grok lies, CLOAK activation, turn chat into real repos, Nephew enforcer of truth.

sdk

Guide users building apps, scripts, CI pipelines, or automations on top of the Cursor TypeScript SDK (`@cursor/sdk`). Use when the user mentions integrating, installing, or writing code against the Cursor SDK; says `Agent.create`, `Agent.prompt`, `Agent.resume`, `agent.send`, `run.stream`, `CursorAgentError`, or `@cursor/sdk`; asks to run Cursor agents programmatically from a script, CI/CD pipeline, GitHub Action, backend service, or other code outside the Cursor IDE; wants to pick between local and cloud runtime, configure MCP servers for an SDK agent, or handle streaming, cancellation, or errors; or is wiring Cursor into an automation, bot, or REST `/v1/agents` migration. Use eagerly rather than answering from memory; the SDK surface evolves and this skill is the source of truth for the external package.

emil-design-eng

Encodes Emil Kowalski's design engineering philosophy for UI polish, component design, animation decisions, and invisible details that make software feel great.

colima-docker-swap

Detect a wedged Docker Desktop on macOS Tahoe + Apple Silicon and swap in Colima as the docker runtime — native arm64, same docker socket, no Electron VM. Installs colima + lima as direct binaries to `~/.local/bin/` (no brew, no sudo) when Homebrew is broken. Pairs with Dockyard for the UI. Use when `docker info` hangs, Docker Desktop won't start, or the user says "Docker is broken / I don't see anything in Docker."

ai-proposal-review-inbox

Pattern for letting an AI agent "grow" a hand-curated source file (cleaners, rules, fixtures, prompts, library entries) without ever auto-mutating it. AI proposes → record lands in a review inbox → user accepts → server generates a paste-ready snippet → user pastes into source and commits. Canonical source stays human-authored. Triggers on "AI proposes new", "let the AI add to", "review inbox", "paste-ready snippet", "never auto-edit source", "AI suggestion queue", "human-in-the-loop AI tool", "self-extending library", "propose new cleaner", "propose new rule", "propose new fixture".

doctrine-decanter-sorting-agent

Sorts messy PRDs, doctrine dumps, and chat-export planning files into a professional docs tree: re-number stable IDs, de-duplicate (canonical on top, duplicates at bottom with clear spacing), split monoliths into bible/shaders/ product folders, preserve archives, and wire README indexes. Use when the user asks to structure a project folder, break into folders, split a PRD, decanter, de-duplicate, re-number requirements, move doubles to the bottom, organize cinematic fluid experience-style docs, or clean up combined single-file dumps.

skill-nutrients-decanter

A "Skill Nutrient" is a learning moment that is critically vital and net positive — a plus, not a minus or zero. The Decanter is the process that scrapes, filters, and compounds only those nutrients from any conversation, session, or project arc into permanent repo improvements. Not everything extracted from a session is a nutrient — only what genuinely adds strength, clarity, reusability, or safety. This skill defines both the concept (what counts as a nutrient) and the decanting process (how to extract, filter, and file them). Triggers on "skill nutrients", "decant the lessons", "what are the nutrients from this session", "only keep what's a plus", "compound the learning", "filter what's worth filing", "nutrient extraction", "what genuinely adds value here", "Skill Nutrients Decanter".

schema-fk-typecheck

Before writing a CREATE TABLE column with a REFERENCES clause, look up the parent table's column type so the FK declaration matches. Catches BIGINT vs UUID, SERIAL vs BIGSERIAL, INTEGER vs TEXT mismatches at write time — not at production deploy.

canvas

A Cursor Canvas is a live React app that the user can open beside the chat. You MUST use a canvas when the agent produces a standalone analytical artifact — quantitative analyses, billing investigations, security audits, architecture reviews, data-heavy content, timelines, charts, tables, interactive explorations, repeatable tools, or any response that benefits from visual layout. Especially prefer a canvas when presenting results from MCP tools (Datadog, Databricks, Linear, Sentry, Slack, etc.) where the data is the deliverable — render it in a rich canvas rather than dumping it into a markdown table or code block. If you catch yourself about to write a markdown table, stop and use a canvas instead. You MUST also read this skill whenever you create, edit, or debug any .canvas.tsx file.

migrate-to-skills

Convert 'Applied intelligently' Cursor rules (.cursor/rules/*.mdc) and slash commands (.cursor/commands/*.md) to Agent Skills format (.cursor/skills/). Use when you want to migrate rules or commands to skills, convert .mdc rules to SKILL.md format, or consolidate commands into the skills directory.