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

Annotations already declare readOnlyHint=true and idempotentHint=true, so the safety profile is covered. The description adds meaningful behavioral context: it states that matching is literal text only and does not interpret a goal, it discloses the return format ({count, results[]}) and the empty-results behavior, and parameter descriptions reveal ordering by name, not relevance, and character stripping. This exceeds the annotation baseline but doesn't cover every possible edge (e.g., no mention of case-insensitivity in the main description, though it is in the query parameter).

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 main description is concise and front-loaded: it states the tool's purpose in the first sentence, then immediately provides usage guidance and sibling differentiation, followed by return format and edge-case behavior. Every sentence earns its place. The parameter descriptions are detailed but not bloated, and the structure groups related guidance logically.

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's complexity (3 optional parameters, an output schema, and 5 siblings), the description is complete. It covers what the tool searches, how matching works, how to narrow by category, how results are ordered, what the return shape is, and what to do instead for other needs. The output schema exists, so return values are already structured; nothing an agent needs to invoke this correctly is missing.

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 coverage is 100%, so the schema documents all parameters thoroughly. The description goes beyond the schema by explaining the matching semantics (literal text only, single terms work best, whole string matched as one substring) and the ordering behavior (by name, not relevance) in the parameter descriptions. It adds practical context like when to raise the limit or use discover_capabilities for ranking. A 4 is appropriate because the description enriches the schema rather than merely repeating it; it's not a 5 because the schema-description coverage is already so high that the description's extra value, though helpful, is incremental.

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 clearly states the tool's purpose: 'Keyword lookup in the Agent Nexus registry when you already know what to look for.' It lists specific searchable fields (product name, vendor, slug, endpoint fragment) and specifies optional narrowing by category. It explicitly distinguishes itself from siblings (discover_capabilities, list_entries, get_entry), making its scope unambiguous.

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 provides explicit guidance on when to use this tool versus alternatives: use discover_capabilities for plain-language needs, list_entries to walk the catalogue in order, and get_entry for a full record of a known slug. It also warns that empty results are a valid answer, not an error, which is a crucial usage guideline. The parameter descriptions add further context, e.g., when to raise the limit or use discover_capabilities for ranking.

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