Upload a file

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

Annotations already declare non-idempotent and non-destructive, but the description adds substance the annotations cannot: a 10 MB size cap, a refusal list (HTML, SVG, XML, scripts, executables), content-type guessing from the extension, and the concrete consequence "Calling it twice stores two files." It also discloses the return envelope, so an agent knows what comes back.

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?

Front-loaded with the verb and resource, then the size limit, then the kind split, then encoding rules and refusals — a sensible order. It is a dense single block with heavy parentheticals, and the trailing return-value sentence is largely redundant given the output schema, but almost nothing else is wasted.

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 six-parameter mutation upload with an output schema, the description covers everything an agent needs to call it correctly: size limit, kind semantics, encoding choice, content-type default, rejected types, attachment target, and duplicate-write behavior. Nothing material is left to inference.

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 a baseline of 3 is warranted, but the description adds meaning the schema lacks: the mutual-exclusivity constraint between text and content_base64 (the schema has no oneOf/anyOf to enforce it) and the default-resolution behavior of content_type when omitted. record_id's single-record limitation is also restated in plainer terms than the schema's regex-scoped field.

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?

Opens with a specific verb and resource ("Uploads one file") and immediately distinguishes the two file kinds it can create, which no sibling does — files_get/list/delete are read or removal operations. The 10 MB ceiling and the two storage modes make the tool's identity unambiguous.

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?

Gives concrete selection conditions: use knowledge_base for documents workers read (playbooks, pricing) versus attachment, and use text for UTF-8 versus content_base64 for binary, "exactly one of the two." That is real when-to-use guidance. It stops short of naming alternatives or stating when not to upload at all (e.g., updating existing knowledge via the knowledge tools).

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