This is the client-side lane for adding Gemini features directly into Firebase apps, whether that's chat, streaming, multimodal prompts, or structured output. Use it when your AI feature lives in the app layer with user-facing UI, not when you need server-owned orchestration like tool calling or RAG. The skill enforces a clean three-way split: app integration here, backend flows in genkit, platform ops in firebase-cli. It won't let you blur those boundaries. The guidance is architecture-first, focusing on where model calls belong, how to wire state and errors, and what Firebase-specific controls like App Check and quota management actually matter at launch. If your request starts client-side but grows into backend territory, it will route you out early instead of pretending one skill owns everything.
npx -y skills add akillness/oh-my-skills --skill firebase-ai-logic --agent claude-codeInstalls into .claude/skills of the current project.
Select a file.
prisma/skills
firebase/agent-skills
Dexploarer/hyper-forge
itsmostafa/aws-agent-skills
prisma/skills