agentfarm_create_runQueues an authenticated asynchronous task. This can consume paid model capacity. Poll with agentfarm_get_run_status, then read agentfarm_get_run_result.
| Name | Required | Description | Default |
|---|---|---|---|
| task | Yes | Task for the agent to complete. | |
| task_input | No | Optional JSON-compatible context or input data for the task. | |
| idempotency_key | No | Optional retry key; the same key and input return the original run. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds valuable behavioral context beyond the annotations: the task is asynchronous, requires authentication, and 'can consume paid model capacity.' This is important cost and execution semantics that an agent needs before invoking the tool. Nothing contradicts the annotations, and the description meaningfully supplements the all-false hint set.
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?
The description is three short sentences with no filler. It front-loads the core action, then adds the cost warning and the follow-up workflow. Every sentence earns its place.
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 an asynchronous create-type tool with no output schema, the description provides enough context to invoke it and proceed correctly: it is async, costs money, and the result is obtained through explicit sibling tools. A small gap is that it does not describe what create_run returns, such as a run ID needed for polling, but the workflow implication is strong enough to keep this near-complete.
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 100%, so the schema already documents task, task_input, and idempotency_key. The description does not add additional parameter-level meaning, but it also does not need to because the schema carries that load. Baseline 3 is appropriate.
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 uses a specific verb and resource: 'Queues an authenticated asynchronous task.' This clearly identifies the action as creating/queueing a run and distinguishes it from the sibling tools that retrieve status, retrieve results, cancel, or return agent info. The follow-up references to get_run_status and get_run_result further reinforce what create_run is not.
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 gives clear workflow context: after creating a run, the agent should poll with agentfarm_get_run_status and then read the result with agentfarm_get_run_result. This effectively tells the agent when this tool fits in the overall lifecycle. It does not explicitly state exclusions or when to prefer alternatives like agentfarm_cancel_run, but the intended usage is clear.
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.
Each tool targets a distinct concern: identity metadata, run creation, status polling, result retrieval, and cancellation. There is no overlap or ambiguity about which tool to call for a given step.
All tools share the agentfarm_ prefix and use a consistent verb_noun pattern (create_run, get_run_status, get_run_result, cancel_run, agent_info). The naming is predictable and uniform.
Five tools form a tight, purposeful surface for managing asynchronous agent farm runs. The set is neither bloated nor too thin for its stated purpose.
The server covers the essential async run lifecycle: create, poll status, retrieve results, and cancel. A list-runs or historical deletion endpoint is missing, but core workflows do not dead-end.