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

Annotations already declare readOnly, non-destructive and open-world, and the description reinforces rather than contradicts them, adding that it returns provenance sources and version history (valuable since there is no output schema). It also discloses an open-world caveat the annotations only hint at: 'One claim does not establish that no other records or sources exist,' and warns against inferring alcohol absence. Much of the surrounding text is refusal policy for the server overall rather than behavior of this call, which caps it below 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 description is dominated by material unrelated to invoking this tool: catalogue scope, refusal scripts for ordering and alcohol guarantees, and offer-price quoting rules. The tool's actual contract arrives near the end, so the text is back-loaded and the agent must read several hundred words before learning what get_claim does. Substantial pruning would improve selection accuracy.

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 single-parameter read tool with no output schema, the description does cover return content (provenance.sources, version history) and open-world limits, so it is not missing critical facts. But the signal is diluted by server-wide policy, and the shape of the returned claim (fields beyond sources) is left vague, so completeness is offset by poor prioritization.

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?

There is one parameter and schema coverage is 0%, so the description must carry the load. It does tell the agent where the value comes from ('Copy claim_uid from a profile or claim search'), which is genuinely useful, but it adds nothing about format, stability, or what happens with an unknown uid. Adequate but thin for the sole required input.

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 tool's function is stated specifically: 'Retrieve a published claim with provenance.sources and version history,' with a clear verb and resource. It also differentiates from siblings, directing agents to use it over a profile claim preview when 'evidence is missing, cropped or contradictory' or a deeper source is needed. The problem is placement rather than content: the purpose sits roughly two-thirds of the way through the description, behind several hundred words of catalogue-scope and refusal policy.

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?

Genuine invocation guidance is present and explicit: use it on a beverage when evidence is missing, cropped or contradictory, or when version history is needed, and 'Do not call it only to repeat a source already returned on evidence.' It also names the alternative (profile claim preview, search_claims) and the input source ('Copy claim_uid from a profile or claim search'). No explicit when-not beyond that one negative case, so it falls short of a 5.

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