parse_document

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

Annotations (readOnlyHint=false, destructiveHint=true, idempotentHint=false) are consistent with the description's disclosed writes to outputPath and credit consumption; no contradiction. The description adds high-value context beyond annotations: exact credit costs per mode, the privacy caveat that extracted content 'may be logged by the host,' and the large-output behavior where spatial lists are kept out of the conversation.

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 deliberately structured with bullet lists for output formats and processing modes, making the cost table and format trade-offs scannable. Every sentence carries decision-relevant information — costs, privacy, output destinations, alternatives — and the slight redundancy with schema credit figures is justified because it consolidates the choice into one place.

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 14-parameter tool with no output schema and zero required parameters, the description covers all invocation-critical aspects: input source constraint, output format destinations, mode costs, privacy/logging implications, and alternative routing to extract_fields. Nothing an agent needs to select a mode, choose a format, or avoid logging sensitive content is missing.

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. The description earns extra credit by explaining what each format actually yields (typed elements with bounding boxes/confidence/reading order vs whole-document Markdown) and by clarifying destination behavior — spatial goes to outputPath, markdown is inline or filed. The cost relationships across formats and modes add meaning the schema's enum descriptions only partially convey.

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 opening line states a specific verb+resource: 'Extract structured data from a document using the Nutrient DWS Data Extraction API.' The body further clarifies what 'structured data' means by enumerating spatial vs markdown outputs. It also distinguishes itself from siblings by naming extract_fields as the alternative for sensitive documents, separating parsing from signer/redactor/processor tools.

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?

Contains explicit decision rules: 'provide exactly one of filePath or url,' 'ask for both up front instead of extracting twice,' and mode selection framed by per-page cost/quality trade-offs. The final note gives an explicit when-not: 'For sensitive documents, prefer spatial output to a file plus targeted extract_fields calls,' and markdown is tied to RAG/search indexing use cases.

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