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

With only three sparse boolean hints in annotations, the description carries the behavioral burden and does substantial work: it discloses the engine implementation (tabula-java lattice/stream with LibreOffice fallback), plan gating on OCR and JSON output ('Starter+'), and a non-destructive inspect capability. The engine fallback chain gives agents an accurate mental model of how 'auto' behaves. No contradiction — readOnlyHint=false aligns with a file-creating extraction tool, and the description's 'non-destructive' claim matches destructiveHint=false.

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?

Roughly 65 words for an 8-parameter tool; the core purpose is front-loaded in the first clause and the remainder is a tight, scannable capability list. The trailing '[category: pdf]' tag and the slightly ambiguous 'non-destructive inspect endpoint' phrase — which could mislead an agent into thinking this tool itself returns inspection results rather than the sibling pdf_to_excel_inspect — are minor blemishes. Every sentence otherwise earns its 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 moderately complex conversion tool with 100% schema parameter coverage but no output schema, the description covers purpose, formats, engine behavior, and feature gates well. The main gaps are return-value semantics (what the agent receives on success: file download URL vs. structured JSON) and explicit failure behavior beyond the schema's OCR-continues-on-failure note. The inspect-endpoint ambiguity also leaves the relationship to pdf_to_excel_inspect slightly unclear, but little is missing for correct invocation.

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 coverage is 100%, so the baseline is 3, but the description adds meaning beyond the schema: it explains the engine implementation (tabula-java) behind the engine enum, flags plan restrictions on ocrFirst and format=json via the 'Starter+' gate, and ties lattice/stream modes to the auto fallback order. It does not deep-document any single parameter, but the added context is actionable for selection and invocation decisions.

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 opens with a specific verb+resource pair — 'Extract tables from PDFs into XLSX / CSV / TSV / JSON' — making the core purpose unambiguous at a glance. It also enumerates distinguishing capabilities (page ranges, sheet modes, OCR, JSON output) that separate this single-file converter from visible siblings like pdf_to_excel_batch and pdf_to_excel_inspect. An agent can identify what this tool does 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?

Usage context is implied through the capability list (engine selection, page ranges, sheet strategy), and the 'non-destructive inspect endpoint' sentence gestures at the inspect sibling, but the description never names alternatives or exclusion conditions. It does not say 'use pdf_to_excel_batch for multiple files' or 'use pdf_to_excel_inspect to validate extraction quality first.' This is implied usage, not explicit when/when-not routing.

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