Google Antigravity rolls out Dynamic Workflows for Python agent scripts

Google rolled out Dynamic Workflows across the Google Antigravity desktop app and CLI on 13 November 2026 for Google AI Pro and Google AI Ultra subscribers, extending a week that added the Gemini 4 Hydrogen Orchestrator and parallel browser subagents. Rather than coordinating long multi-step tasks solely through conversational prompts, invoking /run-workflow instructs the agent to load a built-in workflow skill and write a concise Python orchestration script using parallel() and pipeline() primitives across named phases.

Script approval gate, persistent reviewer agents and JSON Schema checks

Unless an automated execution preset such as Turbo is enabled, Antigravity displays the generated Python script in an approval card for developer inspection before running it in a background process on Antigravity’s bundled Python runtime (or a custom Python 3.8+ interpreter specified via ANTIGRAVITY_PYTHON).

Google Antigravity approval card presenting a generated Python workflow script for developer inspection prior to background execution — Antigravity pauses to show the generated Python workflow script for approval unless an automated preset such as Turbo is active.Image: Google

Workflows can dispatch isolated one-shot workers per file or keep persistent multi-turn agents alive so the same code reviewer later verifies its own requested fixes, validating step outputs against JSON Schemas while keeping verbose worker logs out of the main thread. Approved scripts can be saved in a project and re-run anytime, and Google has not announced standalone API access outside Antigravity.

Technical details

Entry command and surfaces

/run-workflow across the Google Antigravity desktop app and Antigravity CLI

Subscription eligibility

Google AI Pro and Google AI Ultra plans

Orchestration primitives

Python parallel() and pipeline() stages, persistent multi-turn agents and JSON Schema output checks

Python runtime resolution

ANTIGRAVITY_PYTHON override (3.8+), bundled Python runtime, Homebrew, python.org or Windows py

Release details & context

Python interpreter fallback chain and context isolation

When executing an approved workflow script, Antigravity resolves the Python 3.8+ interpreter by checking the ANTIGRAVITY_PYTHON environment variable first, followed by its own bundled runtime and standard system installations from Homebrew, python.org or the Windows py launcher.

Because standard Python control flow handles retries, conditional branches and file batching outside the LLM prompt loop, only structured JSON summaries cross back into the parent chat session, preventing token bloat across repository-wide migrations or multi-pass security audits.

Common questions

Does /run-workflow execute generated Python scripts automatically without asking the developer?

By default Antigravity displays the generated Python script for manual approval first, skipping the prompt only when an automated permission preset such as Turbo is enabled.

Can a Dynamic Workflow reuse the exact same subagent conversation across later pipeline stages?

Yes, scripts can retain persistent multi-turn agent handles so an agent that flags issues in an audit stage can verify the resulting patches in a follow-up stage.

Google release timeline (30 launches tracked)

Earlier11 Nov 2026

Google AI Studio adds self-hosted sandboxes for Gemini Managed Agents

Google enables bring-your-own self-hosted sandboxes for Gemini Managed Agents in Google AI Studio and the Gemini API.

Earlier11 Nov 2026

Google rolls out the Gemini app with hands-free Gemini Live on Apple CarPlay

Google launches the Gemini app globally on Apple CarPlay so iPhone drivers can run hands-free Gemini Live conversations and Workspace briefings.

Earlier12 Nov 2026

Google launches Gemini 4 Hydrogen for everyday coding and agents

Google launches Gemini 4 Hydrogen at $1.50/$7.50 per million tokens, beating Gemini 4 Argon on Artificial Analysis 5.0.1 at 51 and reaching 54 on xhigh.