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

The description clearly discloses that the tool allocates a slot but does not start GPU work until start_training is called, and explains the consent handling behavior. This goes beyond the annotations (readOnlyHint=false, destructiveHint=false) to explain the asynchronous nature and side effects, which is valuable.

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 concise, using three sentences to cover purpose, usage, and behavior. It is front-loaded with the primary action and returns, then adds important caveats. No redundant information.

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 (8 parameters, output schema exists), the description adequately covers the essential behavioral aspects like consent and the non-blocking nature. However, with low schema coverage, more detail on parameters like epochs, fileName, and contentType could improve completeness, but the output schema may clarify return values.

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 coverage is only 25%, with most parameters lacking descriptions. However, the description explains the purpose of acceptsStandingAgreement and how it relates to consent, which is critical for correct usage. It also implies the other parameters (fileName, sizeBytes, etc.) are standard metadata, but could benefit from more detail.

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 tool creates a private NiceVois job and returns a one-time PUT upload URL, distinguishing it from siblings like start_training and list_training_jobs. It specifies the resource (training job) and the action (create), with clear scope (private).

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 explicit guidance on when to use this tool (to create a job before uploading audio) and when not to ask for consent (if standing agreement exists). It also gives a specific flow for handling CONSENT_REQUIRED responses, which is essential for correct usage.

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