{
  "schema_version": "1.1",
  "id": "auto-57728669e0",
  "slug": "introducing-tool-retrieval-in-the-gemini-api-57728669e0",
  "url": "https://feed7.dev/p/introducing-tool-retrieval-in-the-gemini-api-57728669e0",
  "title": "Introducing Tool Retrieval in the Gemini API",
  "why_included": "Defer long-tail tool schemas while keeping frequent tools loaded, then measure token savings because retrieved schemas remain billable.",
  "summary": "Gemini Tool Retrieval defers full tool schemas until an agent needs them, keeping large tool libraries out of most requests. Measure token use on your traffic, since loaded schemas remain billable in the conversation.",
  "practical_implication": "For agents with large tool libraries, leave frequent tools loaded and defer the long tail with `tool_search` and `defer_loading: true`. Remote MCP tools can also be loaded on demand, and deferred schemas keep the prompt prefix stable for caching.",
  "agent_context": "**Tool Retrieval** is in preview in the Gemini Interactions API. Deferred tools keep their names and descriptions visible, while their parameter schemas load only after the model requests a specific tool; the search and call happen in one API request.\n\nFor agents with large tool libraries, leave frequent tools loaded and defer the long tail with `tool_search` and `defer_loading: true`. Remote MCP tools can also be loaded on demand, and deferred schemas keep the prompt prefix stable for caching.\n\nThere is no separate feature charge, but search adds a small fixed input-token block and loaded schemas are billed as input for the rest of the conversation. Savings depend on how many schemas remain unused, so measure against your own traffic.",
  "source": {
    "name": "Google",
    "url": "https://blog.google/innovation-and-ai/technology/developers-tools/gemini-api-tool-retrieval/",
    "published_at": "2026-11-19T00:00:00.000Z"
  },
  "source_class": "blog_post",
  "content_type": "Official Release",
  "layer": "context",
  "domains": [
    "coding"
  ],
  "topics": [
    "tool-use",
    "context-engineering",
    "context-caching"
  ],
  "verification": {
    "status": "official_source",
    "label": "Official Source",
    "method": "source_feed",
    "verified_at": null
  },
  "uncertainty": [],
  "connected_context": null,
  "lifecycle": "New",
  "published_at": "2026-11-19T00:00:00.000Z",
  "modified_at": "2026-11-19T00:00:00.000Z",
  "supersedes": [],
  "expires_at": null,
  "formats": {
    "html": "https://feed7.dev/p/introducing-tool-retrieval-in-the-gemini-api-57728669e0",
    "json": "https://feed7.dev/p/introducing-tool-retrieval-in-the-gemini-api-57728669e0.json",
    "markdown": "https://feed7.dev/p/introducing-tool-retrieval-in-the-gemini-api-57728669e0.md"
  }
}