decide

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

With no annotations, the description carries the full burden, and it delivers: it discloses that the tool performs a 'local forward pass', is 'fast, local, and without hallucination', returns 'calibrated confidence and full probabilities', and includes 'routing metadata' and model auto-selection. It does not detail potential side effects, but the local-inference framing implies a read-only, stateless operation.

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 moderately long but every sentence carries information: purpose, use cases, question-type syntax, return value, and model options are all present and front-loaded. It could be tightened slightly, but there is no redundant or self-evident filler.

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?

The description covers the key call aspects: input shapes for each question type, return payload (confidence, probabilities, routing), and model selection. Since an output schema exists, return-value details are partially covered there. A complete worked example of a full request would be helpful, but the description is sufficient for an agent to call the tool correctly.

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%, so the description must compensate, and it does thoroughly. It explains 'state' as text, email, ticket, or JSON object; defines the 'questions' structure with concrete JSON examples for choice, score, and noul; and documents the 'model' parameter with allowed values and default behavior. Every parameter is given meaningful semantics beyond the raw schema.

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 uses a specific verb ('Answer') and resource ('several typed questions about a state') and clearly differentiates from siblings by emphasizing the batch/composite nature ('several typed questions in one forward pass') versus single-operation siblings like classify/score/check. It also lists the exact judgment types covered (classification, routing, scoring, yes/no gates), leaving no ambiguity about scope.

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 gives explicit context for when to use the tool: 'offload System-1 judgment calls from the main model... fast, local, and without hallucination'. It implies alternatives through the mention of classification, routing, scoring, and yes/no gates, which map to siblings, but it does not state explicit when-not-to-use conditions or name the alternative tools directly.

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