Intent tool: POST /v1/video/queue (settles then enqueues). Returns queue_id + cost. Confirm before calling — this spends prepaid USDC. Poll with get_generation_status or wait_for_video.
| Name | Required | Description | Default |
|---|---|---|---|
| audio | No | ||
| model | Yes | Live video model id from get_models | |
| prompt | Yes | ||
| duration | No | ||
| resolution | No | ||
| aspect_ratio | No |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Beyond the annotations, the description reveals that the call settles a payment, enqueues work, returns queue_id + cost, and consumes prepaid USDC. This is exactly the kind of side-effect and cost disclosure an agent needs before invoking a destructive, non-idempotent tool.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three short sentences, each earning its place: endpoint behavior, return value, and cost warning with polling guidance. The critical cost side effect is front-loaded and nothing is wasted.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The description covers the monetary side effect and the follow-up polling path, but it is incomplete for a 6-parameter tool with no output schema and very low schema coverage. It does not explain how to use the optional parameters or what the required prompt field should contain, leaving a meaningful gap for correct invocation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is only 17%, so the description must compensate for undocumented parameters. It does not: no meaning is added for prompt, duration, resolution, aspect_ratio, or audio, and only the schema itself describes model. An agent invoking this tool still lacks semantic guidance for most inputs.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description names a specific resource and operation: POST /v1/video/queue, and clarifies that it settles then enqueues a video job. It is clear about the tool's role but does not explicitly distinguish it from the sibling video_queue tool.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It gives actionable context: confirm before calling because it spends prepaid USDC, and poll with get_generation_status or wait_for_video afterward. It does not explicitly state when to prefer this tool over its video-related siblings, so it stops short of complete alternative routing.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Several video tools are effectively duplicates: generate_video and video_queue both target POST /v1/video/queue, while get_generation_status and video_retrieve both call POST /v1/video/retrieve. The non-video tools are distinct, but these overlapping boundaries make it hard for an agent to choose the correct variant.
Tool names are uniformly snake_case and many follow a verb_noun pattern such as create_key, list_keys, and get_models. However, the video tools use an object-first video_* pattern, and names like agent_me, chat_completions, and funding_instructions break the dominant convention.
At 18 tools, the surface is on the heavy side, and the count is inflated by lower-level variants that duplicate agent-facing tools such as video_queue vs generate_video and video_retrieve vs get_generation_status. A leaner set could consolidate these while still covering account, key, model, image, and video workflows.
The set covers the account/key lifecycle, funding and price controls, model discovery, chat, image generation, and a full video quote/queue/status/retrieve/cleanup flow. It also provides request-trace recovery and capacity checks, so agents have no obvious dead ends for the stated domain.