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

The annotations already cover safety (readOnly, idempotent, non-destructive), but the description adds real operational context: the agent:read scope requirement, that processing stays on AIProcurement.club, that files are temporary and not retained, and that no Gateway memory, external inference, invoice or idempotency key is used. The hard limits (6 files, 512 KiB combined, 12 pages/PDF, 32k chars/document, 75k/pack) and the untrusted-source-text/OCR-review warning are exactly the kind of disclosure annotations cannot carry. It does not explain the result shape, which keeps it short of a 5.

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?

The purpose is front-loaded in the first sentence and the caveat immediately after, with limits and safety notes following. The limit list is dense and reads as a compressed run-on of units and thresholds, but nearly every clause carries operational or security information rather than filler.

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 file-upload analysis tool with no output schema and two undocumented parameters, the description covers privacy, residency, size caps, OCR behavior, required scope and prompt-injection hygiene. The one notable omission is any indication of what a 'gap' result looks like or how it is structured, which matters because no output schema exists to cover that.

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 has to compensate, and it largely does: it documents the file-count and size ceilings that govern the 'files' array and gives meaning to the 'ocr' enum ('auto handles empty-text pages, always handles mixed image/text', English-only OCR). It says nothing about the id/name/contentBase64 structure of each file object, so the compensation is partial rather than complete.

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 gives a specific verb and resource: 'Identify invoice evidence gaps within supplied files', and it enumerates the gap types (missing/ambiguous order reference, no matching purchase-order document). That is far more informative than the title alone. It never names the sibling it competes with (procurement_invoice__evidence or procurement_purchase_order__evidence), so an agent must still infer the boundary.

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 is implied by the scope caveat 'Does not establish real-world absence', which tells the agent what conclusion it may not draw. However, there is no explicit when-to-use rule and no named alternative for the adjacent evidence/compare/purchase-order tools, so routing between siblings is left to inference.

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