Google Antigravity

#16 of 24
Coding Agents & Harnesses for GTM

▲ 43 60-day move

Measures earned, engagement-weighted share of voice across the GTM voices panel. Methodology →

Google Antigravity positions itself as the command center for agent-first software development: rather than a single coding assistant, it is built as a surface where autonomous agents plan, code, test, and verify work across the editor, terminal, and browser, producing reviewable artifacts like task plans and screenshots along the way. Since its November 2025 debut and the Antigravity 2.0 expansion into a standalone app with CLI and SDK surfaces, it holds a measurable footprint inside the coding-agent conversation, positioned alongside the broader wave of agentic developer tooling. The conversation trades in launch coverage, head-to-head comparisons with rival coding agents, and practitioner walkthroughs of the agent-manager workflow — multiple agents running in parallel across a shared environment.

Measures how easily your agents can build on it — API, MCP, CLI, SDK, docs depth. Methodology →

Antigravity's build-out runs through a Python SDK that abstracts the agentic loop for custom agent development, native support for connecting Model Context Protocol servers (stdio, SSE, or HTTP) alongside built-in tools, a terminal-first CLI, and subagent architecture for hierarchical delegation and background scheduling. The 2.0 release extended this into Managed Agents inside the Gemini API and an enterprise agent platform for organization-wide deployment, moving from a single desktop tool toward a full agent-development stack developers can build on directly.