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

Annotations say readOnlyHint=false (mutation expected) and destructiveHint=false. The description transparently discloses that it may consume credits on first use but free on re-fetch, and includes a 'summary' field as ground truth. It doesn't describe another non-obvious behavior like whether it fails on unprocessed documents, but the provided details exceed the annotations and cover the most important cost and idempotency aspects.

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 sentences, all informative. It front-loads the primary function, clarifies the credit/caching behavior, and directs the agent to use the summary field. No filler or vague marketing language. Slightly more than minimal but not bloated; a well-structured and purposeful description.

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 tool's complexity (credit pooling, caching, summary field, single parameter) and the presence of an output schema (which likely details the returned mappings), the description covers the essential usage context. The description doesn't specify how to handle already processed vs unprocessed documents, but the credit and cache details are likely the main concern. The output schema presumably defines the mapping structure, so this is comprehensive enough.

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?

The input schema has one required parameter `document_id` with zero schema description (0% coverage). The description does not explicitly explain what document_id refers to beyond 'previously processed document'. However, this is a single obvious parameter, and the description's line about 'previously processed document' gives enough context for document_id. Since the schema leaves it undefined, the description partially compensates, but could be more explicit.

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 states the tool runs AI categorization on a previously processed document and returns category mappings. It clearly identifies the verb (categorize/run), resource (document), and distinguishes it from siblings like convert_statement, get_statement, and rate_statement by specifying it returns mappings and handles the categorization toggle.

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 is explicit: says when to use it (after a document is processed), mentions it returns cached results with no charge if already categorized, and explains the credit consumption behavior. It doesn't list alternatives, but sibling context (convert, get, rate) makes the usage context clear. The 'use it as the single source of truth' guidance is also useful for the agent.

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