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

Annotations readOnlyHint=true, openWorldHint=false, destructiveHint=false already signal a safe read/classification operation. The description adds valuable behavioral nuance: the token is opaque, help/correction results lack transactional route tokens, and the classification spans multiple request types. It does not contradict annotations. Minor gap: it doesn't explain what happens when classification is unsupported, but the provided detail exceeds the annotation baseline.

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?

Three sentences are dense but well-organized: purpose first, then classification coverage, then result mechanics. Every sentence earns its place. The parameter guidance in the schema is somewhat long but necessary for correct connectionRef reuse. Slight structural weakness: the classification categories and examples are compressed in the first two sentences, but the overall size is appropriate for the tool's routing role.

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 no output schema, the description compensates by explaining what action results contain (allowed tools, opaque routeToken) and what help/correction results omit. It does not describe the exact shape of the returned classification object (e.g., category field names, confidence), but for a router tool with annotations declaring it read-only and non-destructive, the core decision guidance is present. An agent can safely invoke it and interpret the consequence of the result types.

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 already documents both parameters. The description adds meaningful context about connectionRef: it is opaque, not an API key, contains no credentials, and should be reused across session-id rotations. It also clarifies that 'message' must be the user's complete original message. This goes beyond the schema's terse descriptions, adding behavioral guidance for repeated calls.

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 states a specific verb and resource: classifies a supplied request into help, action, correction or unsupported, and enumerates covered actions (read, create, update, delete, correct, undo, reverse, email, batch). This clearly distinguishes it from the create/update/delete execution siblings: it is a router, not an executor. The help-result behavior further disambiguates it from action routing.

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 explicitly indicates when to use this tool: for classifying an incoming request before choosing an execution tool. It differentiates result types — action results include allowed tools and routeToken, while help/correction results do not issue transactional route tokens. This tells an agent to use this tool first to plan a route, rather than directly calling a create/update/delete sibling, and to route help requests toward brc_find_help_resources or brc_red_help.

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