From google-dev-knowledge
This skill should be used when the user asks about official Google developer documentation, including Android, Firebase, Google Cloud, Chrome Extensions, Flutter, Dart, TensorFlow, Google AI (Gemini), web.dev, Google Ads, or Google Maps Platform. Triggers on questions like "how do I use Jetpack Compose", "Firebase Auth setup", "Cloud Run deployment guide", "Firestore security rules", "Flutter state management", "Chrome extension manifest v3", "TensorFlow model serving", "Gemini API tutorial", or troubleshooting errors from Google services.
How this skill is triggered — by the user, by Claude, or both
Slash command
/google-dev-knowledge:google-dev-docsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Real-time access to official Google developer documentation via the Developer Knowledge API.
Real-time access to official Google developer documentation via the Developer Knowledge API.
| Tool | Purpose | When to Use |
|---|---|---|
search_documents | Search across all supported Google documentation domains | Finding relevant docs by topic, API name, error message, or concept |
get_document | Retrieve full content of a specific document by name | Reading a known document after finding it via search |
batch_get_documents | Retrieve multiple documents in a single request (up to 100) | Deep research requiring content from several related pages |
| Domain | Topics |
|---|---|
| Android Developers | Jetpack, Compose, Architecture, Kotlin, NDK, Play |
| Firebase | Auth, Firestore, Realtime DB, Cloud Functions, Crashlytics, Remote Config |
| Google Cloud | GKE, Cloud Run, BigQuery, Pub/Sub, IAM, Terraform on GCP |
| Chrome Extensions | Manifest V3, APIs, Web Store, DevTools |
| Flutter | Widgets, State management, Platform channels, Packages |
| Dart | Language, Core libraries, Package ecosystem |
| TensorFlow | Keras, TFLite, TF Serving, TFX pipelines |
| Google AI (Gemini) | Gemini API, AI Studio, Model tuning, Embeddings |
| web.dev | Core Web Vitals, PWA, Performance, Accessibility |
| Google Ads | Google Ads API, Campaign management, Reporting |
| Google Maps Platform | Maps SDK, Places API, Routes API, Geocoding |
Documents returned by search_documents include a name field (e.g., documents/firebase--firestore--manage-data--add-data). Use this exact value with get_document and batch_get_documents.
The search_documents tool accepts an optional parent parameter to scope searches to a specific domain. Example: search_documents(query: "auth email link", parent: "firebase"). The documents/ prefix is not needed — pass the domain key directly (e.g., android, firebase, cloud, flutter).
"Jetpack Compose LazyColumn" not "list in Android""Firebase Auth email link sign-in" not "email login""PERMISSION_DENIED Cloud Storage rules""Flutter platform channel invoke method"batch_get_documents accepts up to 100 document names per request. Group related documents for efficient retrieval.
See workflows.md for detailed workflow patterns:
npx claudepluginhub gorban-dev/gor-dev-plugins --plugin google-dev-knowledgeGenerates production-ready GCP code examples from official ADK, Genkit, Vertex AI, and Gemini repos. Adapts to Python/TypeScript/Go with security, monitoring, IaC for AI agents and apps.
Guides WebSearch queries to find authoritative documentation, best practices, implementation patterns, and version-specific details for frameworks, APIs, and databases.
Queries official Microsoft docs via MCP server or mslearn CLI for concepts, tutorials, configuration options, limits, quotas, best practices on Azure, .NET, M365, Windows, Power Platform.