train_start

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

No annotations are provided, so the description carries the full burden. It transparently describes that local training launches a GPU container via docker run --gpus all, pod training ssh-drives pod-native execution, and that the job returns immediately. It also explains post-training behavior: the LoRA is delivered per deliverTo and cataloged when local. This level of detail is excellent for a complex training 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?

The description is information-dense and well-organized, using semicolons to separate major points. While it is relatively long, every sentence adds value, and critical information is front-loaded. It could be slightly more concise without losing clarity.

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?

Given the complexity of the tool (11 parameters, nested objects, no output schema), the description is remarkably complete. It covers the training workflow, target distinctions, job lifecycle, and delivery behavior. It references sibling tools appropriately (train_doctor, train_status, train_cancel) to guide the agent.

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 73%, meaning some parameters lack description. The overall description does not add per-parameter details beyond what is in the schema, but it does mention key parameters like target, pod_id, deliverTo, and datasetPath implicitly. The description adds limited extra semantic value; however, the schema already provides adequate descriptions for most parameters.

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 it starts a LoRA training job, distinguishes between local and pod targets, and mentions the return of a job ID for monitoring via train_status/train_cancel. It also differentiates itself from related tools like train_doctor, train_status, and train_cancel, providing specific context about when to use each.

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 explains when to use local vs pod targets and advises running train_doctor first if unsure about readiness. It implies that the tool returns immediately for long-running jobs and suggests polling train_status. While it does not explicitly state when not to use this tool, the context provided is sufficient for most use cases.

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