Google DeepMind Releases Two New Tools To Keep AI Coding Agents Current With Gemini API
Summary
Google DeepMind releases two new tools — the Gemini API Docs MCP and Gemini API Developer Skills — that together achieve a 96.3% pass rate and 63% fewer tokens per correct answer, keeping AI coding agents current with real-time Gemini API documentation.
Key Points
- Coding agents often generate outdated Gemini API code due to training data cutoff dates, prompting Google DeepMind to release two new complementary tools to address this issue.
- The Gemini API Docs MCP connects coding agents to current Gemini API documentation and SDK information in real time, while the Gemini API Developer Skills provides best-practice instructions and patterns to guide agents toward up-to-date SDK usage.
- When used together, the two tools achieve a 96.3% pass rate on evaluation sets and deliver 63% fewer tokens per correct answer compared to vanilla prompting, with setup available at ai.google.dev/gemini-api/docs/coding-agents.