read_evidence

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

Annotations exist but are notable: readOnlyHint=false on a tool whose description frames it purely as a fetch of an image and an ID, which implies possible side effects (e.g. evidence creation/caching) that the description never explains. The description does add real behavioral value by defining the coordinate contract and orientation caveat, but it says nothing about permissions, limits, or what side effects accompany the call.

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 front-loaded: the return value comes first, then the coordinate semantics, then the orientation warning. Every clause earns its place; it is slightly terse given the tool's read/return nature, but there is no 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?

With no output schema, the description correctly states what is returned (page image and evidence ID), and it covers the non-obvious coordinate formats for both required parameters. Remaining gaps are minor: no mention of side effects hinted at by readOnlyHint=false, no sibling differentiation.

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 must carry the parameter burden, and it does: region.page is 1-based and box is a normalized [left,top,right,bottom] on the original page, with the [0,0,1,1] full-page convention explained. It does not mention the page maximum of 20 from the schema, but document_id is self-evident.

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?

States a specific verb+resource: returns the original-page evidence image plus an evidence ID. That clearly separates it from generic document reads like read_document or read_element. However, it never distinguishes itself from the very similar sibling read_element_evidence, leaving the agent to guess which evidence reader to pick.

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?

Provides procedural guidance for the region parameter (start with [0,0,1,1] for the full page, then narrow down) and a warning not to reuse OCR coordinates of unknown orientation. That is useful, but there is no explicit when-to-use / when-not-to-use guidance relative to the alternatives (read_element, read_element_evidence, crop_document).

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