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

Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, openWorldHint=false. The description additionally discloses that the tool returns only identities (not full content) and explicitly lists excluded fields (content chunks, embeddings, scores, prompts, arbitrary metadata). It also specifies the 'at most 5' cap, which is behavioral context not in the annotations. No contradictions; the description adds meaningful behavioral detail beyond the hint flags.

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?

Single, tightly scoped sentence. Every clause delivers value: the scope ('published support documents for one organization-owned Supovia website') and the output cap with exclusions ('Returns at most 5 safe document identities; content chunks, embeddings, scores, prompts, and arbitrary metadata are excluded'). Nothing redundant or wasted.

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?

Given the tool has an output schema (not shown but referenced), the description needn't explain return structure beyond noting the identity-only shape and the cap. It clearly states the search domain, the filter (organization-owned website), and the output limitation. For a search tool with two required parameters and one optional locale, this is complete for correct invocation.

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?

Input schema already covers 100% of parameters with descriptions (query, locale, websiteId). The description adds no new parameter-level details but reinforces that the search is over 'published support documents' and 'organization-owned website', which clarifies the semantics of websiteId and the scope of the query. Since high schema coverage grants a baseline of 3, the description's confirmation of scope elevates it to 4.

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?

Description clearly states a specific verb ('search') and resource ('published support documents'), scoped to 'one organization-owned Supovia website'. It explicitly limits the return to 'at most 5 safe document identities' and excludes content chunks, embeddings, scores, prompts, and arbitrary metadata, which sharply differentiates it from list_documents (which presumably returns all documents) and get_document (which likely returns full content). This is precise and distinguishes the tool from siblings.

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?

The description implies usage from the singular path ('Search published support documents') and the restriction to 'organization-owned' websites, but does not explicitly contrast with alternatives or state when not to use it (e.g., when full document content is needed). Since siblings like get_document and list_documents exist, the description could indicate 'when you want a filtered list of matches, use this; when you need full content, use get_document'. Still, the scope and the 'safe identities only' limitation effectively route an agent toward the right context.

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