google-devdocs_skill

This skill retrieves up-to-date Google developer documentation via the Developer Knowledge API, delivering Markdown and precise answers from trusted sources.
  • Python

0

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill spm1001/trousse --skill google-devdocs

  • SKILL.md4.9 KB

Overview

This skill orchestrates lookups of official Google developer documentation using the Developer Knowledge REST API. It performs a two-step search-then-retrieve workflow (curl + jq) to return fresh Markdown from 11 indexed Google developer sites without scraping. Use it when you need current, authoritative docs for Cloud, Firebase, Android, TensorFlow, Gemini, and related Google developer platforms.

How this skill works

First, it issues a natural-language search query to the Developer Knowledge API to return up to 10 document chunk snippets and parent document names. Read those snippets to see if they answer the question. If more context is required, fetch the full document Markdown by retrieving the parent document or batch-getting up to 20 documents. All network calls require the GOOGLE_DEVKNOWLEDGE_API_KEY environment variable; if it’s missing the skill stops and informs the user.

When to use it

  • User asks about Google Cloud, Firebase, Android, Chrome, Gemini, TensorFlow, or web.dev APIs
  • You need up-to-date official docs because training data may be stale
  • Troubleshooting a Google API error by searching the exact error message
  • Comparing Google services (e.g., Cloud Run vs Cloud Functions)
  • When you want Markdown-formatted authoritative documentation instead of scraped HTML

Best practices

  • Always check that $GOOGLE_DEVKNOWLEDGE_API_KEY is set before making calls
  • Start with search snippets — they often answer the question and avoid unnecessary full fetches
  • Use natural-language queries (e.g., 'How to authenticate with Firebase Admin SDK') rather than terse keywords
  • Fetch full documents only for the most relevant parents and keep batch retrievals to three or fewer when possible
  • Handle 403/404/429 responses explicitly and report actionable next steps to the user

Example use cases

  • Find the current example for authenticating to Firestore with the Admin SDK
  • Search Cloud Run error text to find recommended fixes in official docs
  • Retrieve the full Markdown guide for Android background work best practices
  • Compare configuration instructions for Cloud Build vs Cloud Deploy
  • Fetch the latest Gemini developer guide section on model invocation

FAQ

The skill stops and informs you to set GOOGLE_DEVKNOWLEDGE_API_KEY; it will not attempt API calls without the key.

Can I restrict results to a single site?

The API has no site filter. Narrow results by using precise natural-language queries like 'Firebase authentication' or include site-specific terms in your query.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational