\n<\/p>\n
In this configuration:<\/p>\n
security-gate<\/code> group runs gate.py<\/code> before every code_execution<\/code> or write_file<\/code> call. If the script returns {\"decision\": \"deny\", \"reason\": \"...\"}<\/code>, the tool call is skipped and the rejection reason is passed into the model\u2019s context.<\/li>\n- The
auto-format<\/code> group runs auto_lint.py<\/code> after every tool finishes to enforce code styling.<\/li>\n- Hooks also support
http<\/code> type handlers that POST directly to an external endpoint.<\/li>\n<\/ul>\nFor complete HTTP hook definitions and failure handling semantics, refer to the hooks documentation.<\/p>\n
Teams are already using hooks to build production-grade validation pipelines. For example, AI-native investment bank Offdeal uses post_tool_execution<\/code> hooks to run automated image verification inside the remote sandbox.<\/p>\n“OffDeal is an AI-native investment bank, and Archie is the AI analyst our bankers use every day. A requirement for banker-ready decks is company logos: buyer tables, sponsor columns, tombstone grids, often 30+ logos in a single deck, every one of which must be the right company, the appropriate size and aspect ratio, contain the name, have a transparent background, and have a high contrast when placed on a white slide.<\/i><\/p>\n
Before agent hooks, we couldn\u2019t do this on Gemini\u2019s managed agents: the sandbox is remote, so our validation code had nowhere to run. With hooks, a post_tool_execution hook triggers our pipeline inside the sandbox the moment Archie writes its company list, fetching candidates, enforcing pixel-level quality checks, verifying each logo with Gemini vision, and publishing a manifest of approved files that are the only images allowed into the deck.”<\/i><\/p>\n
– Alston Lin, Founder & CTO of OffDeal<\/p>\n
Cost control and automation features<\/h2>\nFree tier availability<\/h3>\n
Managed agents are now available on free tier projects. Developers can experiment with agentic workflows using an API key from a project without active billing.<\/p>\n
Budget controls<\/h3>\n
Because managed agents execute multi-turn autonomous loops, complex tasks can consume significant token budgets. To prevent runaway tasks, you can pass max_total_tokens<\/code> inside agent_config<\/code> to cap total consumption (input + output + thinking).<\/p>\nWhen the agent reaches the limit, execution safely pauses and the interaction returns status: \"incomplete\"<\/code>. The environment state is preserved, enabling you to continue where it stopped by passing previous_interaction_id<\/code> with a fresh budget.<\/p>\n<\/div>\n