Skip to main content
Glama

bg-antigravity

Destructive

Run Antigravity CLI as a detached background task for long-running autonomous work. Start a job, poll status, tail logs, and stop or remove tasks without blocking the MCP call.

Instructions

Run Antigravity CLI as a detached background task. Subactions: start (returns taskId), status, tail (log output), stop, remove, list. Use for long-running autonomous work — the MCP call returns immediately and you poll status later.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "action": "<action>"
    },
    "name": "bg-antigravity"
  }
}

Input Schema

NameRequiredDescriptionDefault
cdNo
modeNoAgent mode (--mode): accept-edits, auto (sandboxed autonomy; not available on every account), or plan (answers go into a plan artifact).
labelNoHuman-readable label to identify this task later.
linesNoLines to return from log tail.
modelNoOptional model override. Known: Gemini 4 Hydrogen (Low), Gemini 4 Hydrogen (Medium), Gemini 4 Hydrogen (High), Gemini 4 Hydrogen (XHigh), Gemini 4 Argon (Low), Gemini 4 Argon (Medium), Gemini 4 Argon (High), Gemini 4 Carbon (Low), Gemini 4 Carbon (Medium), Gemini 4 Carbon (High), Orchestrator (Argon), Orchestrator (Carbon), Gemini 3.8 Flash (Low), Gemini 3.8 Flash (Medium), Gemini 3.8 Flash (High), Claude Opus 5.5 (Low), Claude Opus 5.5 (Medium), Claude Opus 5.5 (High), Claude Sonnet 5.5 (Low), Claude Sonnet 5.5 (Medium), Claude Sonnet 5.5 (High), GPT-OSS 120B (Medium).
actionYesBackground task action.
effortNoReasoning effort (--effort). Supported levels depend on the model — see list-models. Use with a base model ID (e.g. gemini-4-hydrogen); not supported by Orchestrator.
promptNoPrompt (required for action=start).
signalNoSignal for action=stop (default SIGTERM; SIGKILL for a run that ignores it).SIGTERM
taskIdNoTask ID returned by action=start.
addDirsNo
sandboxNo--sandbox: run with terminal restrictions enabled.
timeoutNoagy --print-timeout in milliseconds. Omit for no limit (agy waits for the turn and its background tasks).
workingDirNo
skipPermissionsNo--dangerously-skip-permissions. Default: true (print mode can't answer permission prompts), false when `mode` is set. Not allowed with mode plan.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.3.0

TDQS

A3.6/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare destructiveHint=true and openWorldHint=true, and the description usefully adds the asynchronous lifecycle (immediate return, later polling, taskId handoff). However, it never says what remove deletes or what stop does to in-flight work — the two destructive actions — and gives no indication of what status/list return. That leaves the destructive surface under-described given the annotation.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three short sentences, front-loaded with the core action and followed by the action list and the async rationale. No filler, though the action enumeration and the usage sentence partially overlap in intent.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 15-parameter tool with no output schema, the definition covers the action lifecycle but not the return shapes for status, tail, or list, which an agent must poll and interpret. It also omits any mention of how the task's lifecycle interacts with the destructive remove/stop actions, leaving real gaps at this complexity level.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 80%, so most parameters are already documented in the schema (mode, model, effort, signal, skipPermissions constraints). The description adds only the parenthetical "start (returns taskId)", which links taskId back to start; it says nothing about cd, workingDir, addDirs, or label, the parameters the schema leaves bare.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ("Run Antigravity CLI as a detached background task") and enumerates the six subactions, including what start returns. It implicitly separates itself from synchronous siblings by emphasizing "detached" and "returns immediately," but never names ask-antigravity, delegate-antigravity, or batch-antigravity to make the boundary explicit.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

"Use for long-running autonomous work — the MCP call returns immediately and you poll status later" gives a clear condition that selects this tool over synchronous alternatives. It stops short of stating when NOT to use it or naming the sibling to use instead for short/interactive work.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.