template/SKILL.md
Use when [describe triggering conditions, symptoms, or specific errors].
npx skillsauth add guanyang/antigravity-skills replace-with-skill-name-using-hyphensInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
[1-2 sentences explaining the core principle. What is the fundamental insight of this skill?]
[other-skill-name] instead.[Explain the mental model. Use a before/after code block or a small dot flowchart if the decision process is non-obvious.]
// ❌ Anti-pattern / Baseline failure
[Show the common mistake or naive approach]
// ✅ Best practice / Correct implementation
[Show the correct pattern]
[Clear, numbered steps for the agent to follow. Use imperative verbs.]
[A table for fast lookup of commands, syntax, or common operations.]
| Operation / Scenario | Syntax / Command | Notes |
|----------------------|------------------|-------|
| [Operation] | [Command] | [Note] |
[How to prove the implementation is correct (The GREEN phase). Agent must run these checks.]
[List specific rationalizations, shortcuts, or misunderstandings the agent might have. Address the "RED" phase baseline failures here.]
| Red Flag (Excuse / Mistake) | Reality / Fix (What to do instead) | |-----------------------------|------------------------------------| | "I'll just skip testing this time." | STOP. You must verify the behavior first. | | [Mistake or excuse] | STOP. [Correct action] |
tools
Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).
data-ai
Upgrade Remotion, its related packages, compatible Mediabunny packages, and installed Remotion Agent Skills. Use when asked to upgrade or update a Remotion project.
documentation
Search and fetch Remotion documentation pages
tools
Best practices for Remotion