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

Annotations establish write/non-idempotent/openWorld, and the description adds meaningful side-effect detail beyond them: it stores the document and result, bills one credit per five pages for subscription workspaces, applies a 30-day retention window and workspace storage limit, and warns against auto-retrying after a timeout. This is exactly the supplementary context annotations cannot carry.

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 purpose and the file-source preference, then credits/BYOK, then retry policy, so scanning order matches decision order. It is dense and mixes several concerns in single sentences, but almost every clause carries operational weight.

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 write tool with nested objects and an output schema, it covers input sourcing, size and format limits, billing gates, retention, and retry semantics, and correctly closes with a pointer to review confidence and validation errors rather than restating return fields.

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 0%, so the description carries the load and does well: it explains file vs file_inline exclusivity, canonical base64 from exact user bytes with original file_name and matching mime_type, the 32 KiB cap, and the agreement preconditions for use_existing_credits and confirm_provider_charges. schema_slug is left unelaborated, and the supported-format list (PDF/JPEG/PNG/WebP) omits the image/tiff value present in the schema enum.

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 first sentence gives a specific verb and resource ('Extract an authorized attached PDF or image using a saved Dokyumi schema'), which distinguishes it from retrieval siblings like dokyumi_get_extraction. It never explicitly contrasts with those siblings, but the action and input domain are 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?

Explicit routing rules: prefer host-provided download_url/file_id, fall back to file_inline only when unavailable, supply exactly one of the two, run preflight before a BYOK run, and reuse request_key on retry instead of starting a new extraction. Prerequisites for credits and provider charges are spelled out with the required user agreement step.

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