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

Annotations already declare this a read-only, idempotent, non-destructive, closed-world read; the description goes well beyond them by naming the exact deliverables and the tri-state check semantics (PASS, MISMATCH with difference, NOT CHECKABLE). It also scopes the tool negatively — it does not pay, approve, or send invoices — and discloses the no-OCR limitation that will cause silent failures on scanned files.

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 triggering condition, then outputs, then check semantics, then limits and exclusions. The long field list in parentheses is dense but each item earns its place by helping an agent match user requests. No filler sentences.

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?

Though an output schema exists, the description still usefully characterizes the return shape and the meaning of check results. For a one-parameter extraction tool with hard input limits and a non-obvious failure mode (scanned PDFs), nothing an agent needs to call it correctly 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?

The single parameter is fully documented in the schema (100% coverage), which already sets minItems/maxItems of 1-20. The description adds meaning the schema lacks: the files are invoice PDFs in any order and must be text-based, under 10 MB and 20 pages. It does not explain the download_url/file_id mechanics, but the schema covers that.

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?

Opens with a specific verb and resource (extract/check invoice PDFs) and enumerates the exact fields extracted, so an agent knows precisely what it gets. The output artifacts (Excel workbook, CSV ZIP, per-invoice arithmetic checks) further pin down what the tool is versus the sibling check_invoice_math.

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?

States when to use (1-20 attached invoice PDFs, spreadsheet needed, data extraction, arithmetic checking) and explicitly when not to use (phone-photographed receipts, creating a new invoice). Input constraints — text-based PDFs only, no OCR, 10 MB and 20 pages per file — are stated up front rather than left to error handling.

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