Index of every page on this site: llms.txt
All skills
google-antigravity avatar

/google-antigravity-sdk

@97c5c53
Run it yourself

Design, implement, and debug autonomous AI agents and multi-agent systems using the Google Antigravity (AGY) SDK. ACTIVATE this skill when the user wants to create, configure, or orchestrate Google Antigravity agents.

Use this Skill: https://skilld.dev/gh/google-antigravity/antigravity-sdk-python/google-antigravity-sdk

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈60 tokens for metadata: the name and description. ≈2.1k when used: this file. ≈25k more on demand in 27 files.

Description uses 3.0% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈242 characters, or 3.0%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Google Antigravity SDK

Installation & Setup

Before proceeding with any Google Antigravity tasks, ensure the environment is ready:

  • Verify Applicability: If operating in an existing codebase, verify that using this Python SDK is possible and appropriate for the project.
  • Check Dependencies: Check if google-antigravity is listed in the project's dependencies (e.g., requirements.txt, pyproject.toml).
  • Install Package: Ensure the google-antigravity Python package is installed.
  • Authentication Setup:
    • The SDK defaults to hosted Gemini models with an API key (LocalAgentConfig). When running on-device or without cloud connectivity is desired, local models (LiteRTAgentConfig or LocalOpenAIAgentConfig) can be used as an alternative without an API key or cloud credentials.
    • Hosted Models (Gemini - Default): Check for a valid GEMINI_API_KEY environment variable or a .env file (required to access Gemini models).
      • If credentials are missing, you MUST actively help the user get set up with an API key by providing the following link:
        • Default to Google AI Studio: https://aistudio.google.com/app/api-keys
      • Explain that the API key can be passed explicitly in code as shorthand (e.g., LocalAgentConfig(api_key="...")) or automatically read from the environment.
      • For Gemini Enterprise Agent Platform (formerly Vertex AI) authentication, the SDK supports both Standard Mode and Express Mode:
        • Standard Mode (ADC): Instruct the user to run gcloud auth application-default login and configure the agent with vertex=True along with project and location in LocalAgentConfig.
        • Express Mode (API Key): Configure the agent with vertex=True along with api_key="your-express-api-key" in LocalAgentConfig (no ADC or regional project/location needed).
    • Local Models (Alternative): For local models (LiteRTAgentConfig or LocalOpenAIAgentConfig), no API key or cloud credentials are needed. LiteRT is the supported on-device runtime for local models (such as Gemma 4 26B). See references/local_models.md and examples/getting_started/local_models.md for setup details.

Routing Table

Use the following information to dig deeper into specific topics based on the user request. Read the referenced files or explore the directories to find relevant information.

References

  • If the user needs to understand the high-level overview and core concepts of the Google Antigravity SDK (Agent, Conversation, Connection), read references/architecture.md.
  • If the user needs to perform advanced agent configuration (e.g., selecting appropriate models, configuring execution behavior via agent_behavior—defaulting to autonomous vs interactive—or configuring connection reliability), or understand the critical rules for model identifiers to avoid assumptions, read references/agent_configuration.md.
  • If the user needs to extend an agent's capabilities by integrating Model Context Protocol (MCP) servers, or configure tool permissions for the agent, read references/mcp_integration.md.
  • If the user needs to define safety policies, resolve execution order, restrict agent actions using predicates, or run terminal commands inside an OS-level sandbox, read references/safety_policies.md.
  • If the user needs to debug failed agents, stream logs, or implement error recovery using hooks to make agents robust, read references/error_handling.md.
  • If the user needs to monitor costs, track token usage (including thinking tokens), or build custom audit logs for advanced monitoring, read references/observability.md.
  • If the user needs to see a list of built-in tools and understand their default state, read references/built_in_tools.md.
  • If the user needs to run agents locally using on-device models (LiteRTAgentConfig for the supported on-device runtime, or LocalOpenAIAgentConfig for external OpenAI-compatible servers like Ollama/LM Studio), understand hardware requirements, or configure local execution, read references/local_models.md.

Examples

  • If the user needs to implement basic agent behavior, streaming responses, or expose internal thoughts, read examples/getting_started/hello_world.md.
  • If the user needs to customize or override default retry behavior and exponential backoff for API errors or schema validation, read examples/getting_started/customizing_retries.md.
  • If the user needs to equip an agent with custom capabilities (tools) derived from Python functions, or maintain agent state across tool execution, read examples/getting_started/custom_tool.md.
  • If the user needs to shape an agent's persona, define its system instructions, or dynamically adapt its behavior, read examples/getting_started/persona_config.md.
  • If the user needs to build multimodal agents capable of processing images and PDFs, or generating visual content, read examples/getting_started/multimodal.md.
  • If the user needs to implement multi-agent delegation, allowing a main agent to spawn and orchestrate subagents, or configure multi-tier nested subagent hierarchies (using max_subagent_depth and allowed_subagents), read examples/getting_started/subagents.md.
  • If the user needs to author and execute deterministic multi-agent workflows using @beta.workflows.define (phase, log, agent, parallel, pipeline) and await agent.beta.run_workflow(...), read examples/getting_started/workflows.md.
  • If the user needs to connect an agent to external services via MCP (Stdio or SSE), read examples/getting_started/mcp_tools.md.
  • If the user needs to create proactive agents that respond to time-based events or file system triggers in the background, read examples/getting_started/periodic_trigger.md.
  • If the user needs to intercept agent lifecycle events (e.g., pre/post turn, stop, tool execution, errors) to customize execution flow, read examples/getting_started/hooks.md.
  • If the user needs to implement turn-level cancellation or programmatic stream aborts, read examples/getting_started/cancellation.md.
  • If the user needs to implement persistent agents that remember past interactions across sessions, read examples/getting_started/persistence.md.
  • If the user needs to override the default application data directory for agent artifacts, scratch files, and media storage, read examples/getting_started/app_data_dir_override.md.
  • If the user needs an agent to output structured data (e.g., JSON matching a Pydantic schema) for reliable integration, read examples/getting_started/structured_output.md.
  • If the user needs to add, configure, or load agent skills into the Google Antigravity SDK agent, read examples/getting_started/agent_skills.md.
  • If the user needs to enable and use built-in web tools (like Google Search or URL fetching) with the agent, read examples/getting_started/web_tools.md. (Note: when fetching massive web pages or articles, pair read_url_content with view_file to inspect cached disk files).
  • If the user needs to enforce session operational limits (model or tool calls) or proactive token budget controls (input, output, or total tokens) and handle StopReason, read examples/getting_started/budget_limits.md.
  • If the user needs to set up and run a local model agent (LiteRT, or an OpenAI-compatible server like Ollama), including model download, hardware requirements, and context compaction configuration, read examples/getting_started/local_models.md.
  • If the user needs to configure conversation context limits and compaction thresholds to handle long-running sessions, read examples/getting_started/compaction.md.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 alerttoday3 checks · Risk SAFE
  • Gen Agent Trust Hub1d

    This skill provides comprehensive documentation and usage examples for the Google Antigravity SDK. It describes legitimate features for building AI agents, including tool use, local model execution, and multi-agent workflows, while emphasizing security best practices such as safety policies and sandboxing.

  • Sockettoday

    No alerts

  • Snyk1d

    Risk: CRITICAL · 3 issues

Signed by skilld at 97c5c53. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated 2 months ago

README badge

README badge for google-antigravity/antigravity-sdk-python