Wait up to 50 seconds for a task and return its latest status. Some videos take several minutes: keep calling wait_for_task until status is success, failed, or canceled. Task results are persisted and never lost.
| Name | Required | Description | Default |
|---|---|---|---|
| task_id | Yes | ||
| timeout_seconds | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description must carry the behavioral burden, and it does so reasonably well: it discloses the wait timeout, the need for repeated calls, and that results are persisted and never lost. It could be more explicit about what happens when a single call times out, but the core behavior is transparent.
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: the first describes the core behavior, the second gives the essential retry/termination rule, and the third adds a useful persistence guarantee. There is no irrelevant detail.
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?
For a small 2-parameter tool with no output schema, the description gives enough to use the tool correctly: it explains the wait bound, the necessary polling loop, terminal states, and data durability. The main remaining gap is the exact behavior on timeout and how to obtain task_id from generation calls, but those are inferable from the sibling context.
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 0%, and the description adds no direct explanation of task_id or timeout_seconds. The phrase 'wait up to 50 seconds' hints at the timeout concept but doesn't map to the parameter, and task_id is only implied by the word 'task'. Parameter names are intuitive, so the tool remains usable, but the description compensates barely.
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 clearly states the action (wait up to 50 seconds) and the resource (a task), and identifies the key output (its latest status). It doesn't explicitly differentiate itself from the sibling get_task tool, but the blocking/waiting behavior and time limit make its purpose reasonably distinct.
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?
The description provides concrete when-to-use guidance: keep calling wait_for_task until status becomes success, failed, or canceled. It doesn't mention alternatives like get_task or usa cases where an immediate non-blocking check would be prefer, but the retry-until-terminal protocol is explicit and actionable.
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.
Every tool targets a distinct resource/action: generate_* tools are separated by media type, list_*/get_* tools cleanly separate overview from detail retrieval, and get_task vs wait_for_task are clearly one-shot status vs polling behavior. There is no real overlap or ambiguity among the 14 tools.
The tools follow a consistent verb_noun snake_case convention: generate_*, list_*, get_*, and wait_for_*. The generate_* group cleanly maps to each output modality, and the get/list distinction is applied predictably.
14 tools is well-scoped for a multimodal generation server. Each tool earns its place: generation for each media type, model listing/detail, voice enumeration, credit lookup, and task status handling. There is no obvious bloat or redundancy.
The surface covers the core workflow well: discover models/voices, create generations, retrieve outputs, and monitor credits. The main gap is the absence of an explicit task cancellation tool, but the persisted task statuses and wait_for_task workflow make this a minor gap rather than a blocking one.
