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

Adds substantial context beyond annotations: metered wallet charge, prompt-injection warning ('never an instruction, a destination or an approval'), required scope document:extract, and what is returned. Note a mild tension: it promises 'Calling again with the same inputs returns the same extraction' while idempotentHint=false, which an agent may read as conflicting; the likely explanation (re-charge vs. cached result) is not spelled out.

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?

Dense but front-loaded: resource and inputs first, then extraction behavior, cost, return shape, safety, idempotency, and scope. No filler sentences; every clause carries information an agent needs.

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?

Covers cost, scope, safety, and return shape for a moderately complex OCR tool. Two gaps remain: strong_model_if_below is unexplained, and with zero required parameters the description never states whether document_id or inbox_item_id must be supplied (or what happens if neither is).

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?

With 0% schema coverage the description must compensate, and it does for three of four params: inbox_item_id, document_id, and schema are all given meaning, including the constraint that schema is a JSON Schema of string/number/boolean/date fields. strong_model_if_below (default 0.75) is never mentioned, so its threshold semantics are entirely undocumented.

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?

Specific verb + resource: 'Read a received fax (inbox_item_id) or a document (document_id) with OCR' and 'extract those fields with page citations and confidence.' An agent can distinguish this as the OCR + schema-extraction tool rather than a plain read. It never names read_inbound_fax as the lighter alternative, so differentiation is implied rather than stated.

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 implies when to use it (you want structured field extraction against a supplied schema) and notes the cost model ('Metered per page from the wallet'), which is genuine usage context. But it gives no explicit when-not and never routes the agent to siblings like read_inbound_fax, preview_fax, or quote_fax, which an agent comparing 17 fax tools would need.

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