Google Antigravity SDK 0.1.22

Changelog: Dynamic workflow orchestration, in-memory skill definitions, and configurable per-step output token limits

Google Antigravity Python SDK v0.1.22 introduces beta support for dynamic workflow orchestration, in-memory skill definitions, and configurable per-step output token limits for extended model decodes. This release also standardizes subagent invocation naming on BuiltinTools.INVOKE_SUBAGENT and provides fine-grained controls over MCP tool payload limits and eager tool registration.

Improvements(17)

• Dynamic Workflow Orchestration (beta.workflows): Author and execute multi-step agent workflows with typed workflow primitives (phase, log, agent, parallel, pipeline), the @beta.workflows.define decorator, and agent.beta.run_workflow().
• In-Memory Skills (InlineSkill): Define and register in-memory skills directly on agent configurations using InlineSkill, eliminating the requirement for disk-based skill files.
• Configurable Output Token Limits (GeminiModelOptions): Configure per-step output generation limits via max_output_tokens in GeminiModelOptions to support long decodes exceeding 64k tokens.
• Canonical Subagent Tool Naming (BuiltinTools.INVOKE_SUBAGENT): Standardize subagent invocation around BuiltinTools.INVOKE_SUBAGENT, deprecating BuiltinTools.START_SUBAGENT while retaining automatic coercion and dictionary lookup compatibility.
• Configurable MCP Output Byte Limits (CapabilitiesConfig): Configure byte thresholds for Model Context Protocol (MCP) tool responses via mcp_max_output_bytes on CapabilitiesConfig to handle large payloads without truncation.
• Eager MCP Tool Loading (force_all_tools_eager): Opt out of lazy tool-search loading to force all tools from an MCP server into the model's eager tool list.
• Dynamic Workflows: Added the google.antigravity.beta.workflows module with orchestration primitives (phase, log, agent, parallel, pipeline), @beta.workflows.define decorator, AST script extraction, and the agent.beta.run_workflow entry point.
• In-Memory Skills: Exposed InlineSkill at the root package level and added the inline_skills parameter across AgentConfig, LocalAgentConfig, LiteRTAgentConfig, and LocalOpenAIAgentConfig. Passing both skills_paths and inline_skills simultaneously is validated and rejected.
• Subagent Model Configuration: Added support for assigning dedicated model configurations to individual subagents via SubagentConfig(model=...), enabling multi-tier or multi-provider agent swarms.
• SandboxStatus Type Export: Added SandboxStatus to google.antigravity root exports and surfaced is_sandboxed across session contexts.
• UsageMetadata Reflected Subtraction: Implemented __rsub__ on UsageMetadata to support reverse subtraction against numeric zero values.
• Tool Lifecycle Call IDs: Propagated call_id across tool hooks (pre_tool_call, post_tool_call) to correlate execution timing and telemetry traces.
• Workflow Tool Default: BuiltinTools.RUN_WORKFLOW is now enabled by default across standard tool suites.
• Gemini Max Output Tokens: Added max_output_tokens to GeminiModelOptions, allowing models to generate responses up to target token lengths.
• Subagent Invocation Tool Name: Replaced BuiltinTools.START_SUBAGENT with BuiltinTools.INVOKE_SUBAGENT across tool definitions and documentation.
• MCP Lazy Loading Opt-Out: Added force_all_tools_eager to MCP server configuration to disable lazy tool search and load all server tools immediately.
• MCP Output Byte Limit: Added mcp_max_output_bytes to CapabilitiesConfig for customizable tool response payload limits.

Fixes(2)

• Skill Lookup Steps: Server-side skill lookups were misclassified as StepType.AGENT_ACTION instead of StepType.SKILL_LOOKUP in trajectory parsing.
• ToolContext Type Detection: Import aliases ending in .ToolContext (such as import foo as ToolContext) were incorrectly matched as tool context injections.

<h3 class="