agent-antigravity

agent-antigravity is a skill for Claude Code from bgembalczyk/agent-antigravity-mcp. It costs 9 tokens per session (3,740 once invoked), scanned A, original, MIT.

A sandbox agent that can perform delegated tasks, including running code, browsing the web, calling functions, and keeping files between interactions. A sandbox is an isolated working environment.

In plain words
What is it for?
It is for delegating coding, research, web, and multimodal tasks, then checking progress or continuing the same interaction.
Why use it?
It moves substantial work into a separate environment while preserving the task context for follow-up requests.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the agent-antigravity plugin — 1 skill, 1 MCP server shipped together

Good fit It is for delegating coding, research, web, and multimodal tasks, then checking progress or continuing the same interaction.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/bgembalczyk/agent-antigravity-mcp/agent-antigravity
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add bgembalczyk/agent-antigravity-mcp --skill agent-antigravity
Clone the repo
git clone --depth 1 https://github.com/bgembalczyk/agent-antigravity-mcp

Made for: Claude Code.

Or install agent-antigravity, the plugin that ships this one along with the rest of its 1 skill, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for agent-antigravity

README.md
[![agentmods](https://agentmods.dev/badge/skills/bgembalczyk/agent-antigravity-mcp/agent-antigravity/github.svg)](https://agentmods.dev/skills/bgembalczyk/agent-antigravity-mcp/agent-antigravity)
Your own site
<a href="https://agentmods.dev/skills/bgembalczyk/agent-antigravity-mcp/agent-antigravity"><img src="https://agentmods.dev/badge/skills/bgembalczyk/agent-antigravity-mcp/agent-antigravity/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for agent-antigravity

Your own site · 80×15
<a href="https://agentmods.dev/skills/bgembalczyk/agent-antigravity-mcp/agent-antigravity"><img src="https://agentmods.dev/badge/skills/bgembalczyk/agent-antigravity-mcp/agent-antigravity.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 9 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,740 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00009 $0.03740
Opus 5 $0.00005 $0.01870
Sonnet 5 $0.00002 $0.00748
Haiku 4.5 $0.00001 $0.00374

Measured 9d ago against content hash b6c43fffe569, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

agent-antigravity scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/agent-antigravity/SKILL.md · 238 lines

How it starts

The opening of the file, as written. The whole thing — 238 lines — stays where its author put it; the contents beside it link to each section on GitHub.

agent-antigravity MCP server

Delegate real work to this sandbox agent instead of doing it yourself — hand it a clear, complete, self-contained instruction through one of the tools below, then poll or continue the same interaction rather than re-prompting from scratch. It can execute code, browse the web, call functions, and keep sandbox state across turns; treat it like handing a task to a capable but literal-minded coworker who needs the full picture up front, not incremental hints.

Before calling anything, confirm GEMINI_API_KEY is set (the server raises ValueError otherwise); confirm GITHUB_TOKEN too if the task needs create_github_pr_task.

Tool quick reference

Tool Use it to
create_interaction Start a new task: run code, search the web, multimodal input
continue_interaction Send a follow-up turn in the same sandbox (files persist)
submit_function_result Answer a requires_action status from a custom function tool
get_interaction_status Poll a single interaction by ID
list_interactions List cached interactions, newest first
cancel_interaction Stop an in-progress (typically background=True) interaction
delete_interaction Remove a stored interaction and its local cache entry
create_github_pr_task One call: clone repo, branch, do task, push, open PR
create_trigger Schedule a recurring interaction on a cron expression
list_triggers / get_trigger Inspect triggers
update_trigger_status Pause/resume a trigger
delete_trigger Permanently remove a trigger
run_trigger Fire a trigger immediately, outside its schedule
list_trigger_executions Audit a trigger's run history
create_managed_agent Register a reusable agent (persona, tools, environment) by ID
list_managed_agents / get_managed_agent Inspect registered agents
delete_managed_agent Remove a registration
create_environment Provision a standalone sandbox for reuse across interactions
list_environments / get_environment Inspect sandboxes
delete_environment Terminate a sandbox and free its resources
download_environment_files Pull the whole sandbox filesystem as a .tar
upload_environment_file Push a local file into a running sandbox

Read the full file on GitHub · 238 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 238 lines · 9 tokens per session scan A b6c43fffe569

Subscribe to this mod's changes

agent-antigravity is a skill published in the GitHub repository bgembalczyk/agent-antigravity-mcp (2 stars, last pushed 1mo ago), licensed MIT. It adds 9 tokens to every session and 3,740 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.