Skip to main content
Glama

Generate several images in parallel

antigravity_image_swarm

Generate multiple images in parallel by assigning one isolated worker to each prompt. Returns file paths, formats, sizes or per-image errors, with extensions corrected to match real bytes.

Instructions

Best for several images at once: like antigravity_image, but one isolated worker per prompt, run in parallel (capped at max_concurrency). Returns each image's path, format and size, or its error; extensions are corrected to the real bytes. Unsandboxed agy workers: trusted prompts only.

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": {
      "prompts": "<prompts>"
    },
    "name": "antigravity_image_swarm"
  }
}

Input Schema

NameRequiredDescriptionDefault
watchNoIf true, open the live dashboard; each finished image shows in its pane, and clicking a row opens that agent's window beside the dashboard.
promptsYesOne image description per parallel worker.
timeout_sNoPer-worker timeout in seconds. Default 240 (images are slower).
workspacesNoWorking directory per worker (same shorthand as antigravity_swarm).
output_pathsNoWhere to save each image (aligned to prompts). Omit to write timestamped files in the first workspace (or server cwd).
max_concurrencyNoMax workers running at once (default 4).

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.11.0

TDQS

A4.1/5.0
Behavior4/5

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

Annotations already declare a write, open-world, non-idempotent operation, so the description's added value is real: it discloses return content (path, format, size, or per-item error), extension-correction behavior, and an unsandboxed-worker trust constraint that is a genuine safety caveat not captured by the hints. It does not discuss cost or partial-failure semantics beyond 'or its error'.

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?

Two tight sentences that front-load the selection criterion, then the result contract, then the safety caveat. Dense and nearly waste-free, though the 'extensions are corrected to the real bytes' clause is a detail that could sit in the output schema.

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

Completeness4/5

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

For a 6-parameter, non-idempotent, open-world generation tool with an output schema present, the definition covers selection, per-prompt isolation, result shape and the trust caveat, which is enough to call it correctly. The main residual gap is what happens on partial failures when some workers succeed and others time out.

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 100%, so every parameter including max_concurrency, output_paths and workspaces is already documented in the schema. The description only echoes the concurrency cap and the one-worker-per-prompt mapping, adding little semantics beyond the structured fields; baseline 3 is appropriate.

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

Purpose5/5

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

States a specific action (generate several images) and the mechanism (one isolated worker per prompt, run in parallel) while explicitly positioning it against antigravity_image ('like antigravity_image, but...'). An agent can distinguish it from the single-image sibling without opening either schema.

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?

'Best for several images at once' gives clear selection context relative to the single-image tool, and 'Unsandboxed agy workers: trusted prompts only' adds a precondition. It stops short of an explicit 'use antigravity_image when you have one prompt' exclusion, so the routing is implied rather than fully stated.

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