Skip to main content
Glama

Generate assets

generate_assets

Submit one render per format. SPENDS CREDITS (estimate_cost gives the price) unless mock is true. Returns immediately with job tokens; follow with wait_for_jobs. Set max_credits to refuse if the price is higher.

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": {
      "brief": "<brief>",
      "formats": "<formats>"
    },
    "name": "generate_assets"
  }
}

Input Schema

NameRequiredDescriptionDefault
mockNoSimulate with placeholder renders; free
modeNoimage
brandNoThe full brand, only when there is no kit for it. brand_id is simpler.
briefYes
cropsNoStarting crop per format for master sizing, e.g. {"9:16": {"x": 0.4, "y": 0.5}}. Change it later with compose_asset at no cost.
modelNoDefaults: recraft-v4.1 for image, kling-3-std for video. See list_models.
videoNo
sizingNoImages only. master: one render at the native ratio that best covers every format, each format cropped from it (one charge, identical picture in every size). separate: one render per format. Video always renders per format.master
formatsYesSocial formats, see list_formats
max_usdNoThe same cap in dollars ($0.01 a credit)
overlayNo
brand_idNoA brand kit from list_brands (id or name). Brings the colours, fonts, logo, website and default copy (headline, call to action, small print); use instead of brand.
preset_idNoMarketing Studio preset from list_marketing_presets. Needs 1 or 2 references.
referencesNoProduct images or app screenshots for marketing-studio-2 and marketing-studio-flare: public https URLs, studio file paths (uploads/...), or asset ids from the brand kit (get_brand). Product first.
max_creditsNoRefuse (submit nothing) if the price in credits is above this
green_screenNoFor a lifestyle shot of someone using the app: the phone or tablet screen comes out plain bright green, so place_screen can put the real app on it (on the still, or on every frame after animate_image). Describe the camera behind the person (seen from behind their shoulder, looking down at the phone), so the screen can face both them and the viewer; then review_shot the result.
text_overlayNoBake headline, CTA and logo on afterwards in brand fonts (recommended; models misspell text)
negative_promptNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4/5.0
Behavior4/5

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

Annotations only say the call mutates state and is non-idempotent; the description adds the genuinely useful facts that it spends credits, returns job tokens immediately rather than finished assets, and that max_credits causes a full refusal before submission. It does not say whether a failed job is refunded or what happens on partial failure, which matters for a credit-spending call.

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

Conciseness5/5

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

Four short sentences, each doing distinct work: what is submitted, what it costs, what comes back, and the safety valve. The cost warning is capitalized and front-loaded where it will be seen.

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 an 18-parameter async generation tool with no output schema, the description covers the parts an agent cannot infer: async handoff, credit spend, and refusal semantics. The many domain-specific knobs (green_screen, text_overlay, references, sizing) are left entirely to the schema, which is acceptable given 72% coverage but leaves the description thin for such a complex call.

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 72% and the nested objects (brand, brief, crops, video, overlay) are documented in the schema itself, so baseline 3 applies. The description only glosses mock and max_credits, both of which already carry their own schema descriptions, adding no new meaning.

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?

"Submit one render per format" plus the credit/async notes make it clear this is the job-submission entry point for asset generation. It does not distinguish itself from multi-asset siblings like build_campaign_pack or add_scene, so an agent still has to infer that this is the primitive single-batch call rather than a packaged workflow.

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?

It names the two adjacent steps explicitly: estimate_cost for price and wait_for_jobs for the result, and flags mock=true as the free dry-run path. What's missing is negative guidance — when to pick this over build_campaign_pack or compose_asset — but the surrounding workflow is unambiguous.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources