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

Annotations declare readOnlyHint=true and destructiveHint=false, but the description goes far beyond that. It discloses critical behavioral traits: one claim does not imply no other records, confidence is not proof, 'Below 0.5%' is an upper-bound declaration, 0% vs 0.0% wording difference is not proof of differing thresholds, and alcohol absence cannot be inferred. These are non-obvious, high-value constraints that materially affect how an agent interprets the returned data. This is exactly the kind of context annotations cannot provide.

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 first sentence efficiently front-loads purpose and return content. However, the rest is a long list of interpretive warnings that, while valuable, are not clearly prioritized or grouped. The sheer volume of caveats makes it less scannable. It earns its place but could be better structured with bullet points or a clearer separation between core function and interpretation guidelines.

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 read-only, one-parameter tool with no output schema and 0% schema coverage, the description is remarkably complete. It explains what is returned, how to obtain the identifier, and provides extensive interpretive guidance about the returned data (provenance, versions, alcohol wording, confidence caveats). An agent has everything needed to call and correctly interpret the tool.

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% for the single claim_uid parameter. The description compensates by telling the agent where to get the value ('Copy claim_uid from a profile or claim search'). It does not specify the UID format or error behavior if not found, but it covers the essential semantics of the parameter. With only 1 parameter and 0% coverage, this is strong compensation.

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 ('Retrieve a published claim') with details on what is returned (provenance.sources, version history). Distinguishing from get_beverage, get_producer, get_context etc. is implied by the resource name, but there is no explicit sibling routing (e.g., get_context for context, get_entity for entities). It's clear but not maximally differentiated.

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 clear context: 'Copy claim_uid from a profile or claim search' and a use case ('check a source behind a profile claim preview'). This tells the agent when to reach for get_claim rather than search_claims. However, it does not explicitly state when NOT to use it or name alternative tools for adjacent needs (e.g., use get_context when only context is needed).

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