Onboard a new data source, step 1: create a data spec and get upload URL(s)

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

Annotations provide no safety hints, so the description carries the burden. It discloses key behaviors: for file it returns presigned URLs, for tables it uses AI-generated SQL and requires autoRefresh. It also explains the two-step flow. While it doesn't discuss side effects or error conditions, it adds valuable context beyond the schema.

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 two sentences that efficiently convey the core purpose, both modes, the upload URL flow, the next step, and the autoRefresh requirement. It is front-loaded with the purpose and contains no unnecessary words.

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 two modes and autoRefresh, but leaves ambiguity about whether finish_data_source_onboarding is required for tables mode as well, and does not clarify what the tool returns for tables (e.g., a spec ID). Given the tool's complexity (15 params, nested objects, no output schema), it could be more explicit about expected outputs and follow-up steps for both modes.

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 100%, so parameters are well-documented. The description enriches this by clarifying conditional dependencies (e.g., sourceType tables requires autoRefresh, sourceType file requires sampleFileName and yields upload URLs) and explaining autoRefresh values. This adds meaning beyond the schema's individual descriptions.

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 data spec as the first step of data source onboarding, with distinct behaviors for file (presigned upload URLs) and tables (AI-generated SQL). It explicitly references the follow-up step (finish_data_source_onboarding), distinguishing it from sibling tools like update_data_spec or finish_data_source_onboarding.

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 positions this as the first step and instructs calling finish_data_source_onboarding after upload for the file mode, but does not explicitly state when to use this vs. alternatives like update_data_spec. It does clarify the file vs. tables modes and the autoRefresh requirement, providing sufficient context for typical use cases.

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