Process a business document end-to-end

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

Annotations already declare readOnly/idempotent/non-destructive, but the description goes well beyond them: cache key composition and the id_document PII auto-bypass, the isError conditions around OCR confidence, that validation is schema-only (not factual), and the 60000-character rejection behavior. This is unusually rich behavioral disclosure for a read-only processing tool.

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 pipeline summary, then grouped by concerns (sources, kinds, returns, validation, cache, options). Length is defensible for a 3-param nested pipeline tool, but a few statements restate schema facts (the 60000 limit, the 0.3 floor, maxPages/minOcrConfidence scoping), which is mild redundancy given a fully described schema.

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?

With an output schema present, the description still usefully explains the discriminated-union return (switch on kind, extraction_source, ocr_confidence semantics). Combined with the caveats about confidence floors and cache bypass, an agent has everything needed to call and interpret the tool.

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 the baseline is 3, but the description adds framing the schema does not: that maxPages/minOcrConfidence apply only to OCR sources (and how they interact with provided text), that text sources yield null ocr_confidence/page_count, and that the 0.3 floor is a placeholder to calibrate. It adds value beyond the field-level docs, though several details (60000 chars, 0.3 default) are duplicated from the schema.

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 names a concrete pipeline with specific stages (classify → typed extraction → validation) and the input modalities it accepts. It also explicitly positions itself against siblings: 'Replaces the manual chain of mistral_ocr + mistral_chat + JSON parsing.' An agent can distinguish it from mistral_ocr and mistral_chat without opening the schema.

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?

It gives explicit routing guidance: kind=auto lets the server classify, source.type=text skips OCR and Files uploads, text+generic makes no API calls, and 'use mistral_ocr directly if you need raw OCR without a confidence guarantee.' It names the alternative and the condition that selects it.

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