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

Annotations already declare readOnly, idempotent and openWorld, so the safety profile is covered. The description adds real behavioral context beyond them: the index-floor suppression behavior, the rule that a null composite is NOT a zero, and the guarantee that below floors the payload carries counts and a plain sentence rather than a fabricated score.

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?

Front-loaded with the two useful facts (FREE, THE MEASURED RECORD), but the body is a single sprawling sentence that enumerates return fields already covered by the output schema. It is dense and informative yet padded with redundant payload listing, so structure is serviceable rather than tight.

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?

With annotations, a zero-arg schema and an existing output schema, the definition is more than complete for correct invocation; the description needn't restate return values but does. Nothing needed to call it correctly is missing, only some redundancy remains.

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?

The tool takes zero parameters (empty schema, additionalProperties false), so there is no per-parameter meaning to convey; baseline for a no-arg tool is 4. The description correctly signals the tool is a pure read with no inputs.

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 resource (the Decision-Intelligence Index as published at GET /v1/intelligence) and enumerates the exact payload contents — composite, Brier decomposition, evidence counts, price ladder. It is highly specific about what the tool returns, though it never explicitly names a sibling or contrast tool (e.g. tollbridge_calibration, tollbridge_pricing) to help an agent choose among the many tollbridge_* reads.

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?

"Agents buy on this read: 'is the judgment worth the price?'" implies the decision context in which to reach for it, and "Free read — no payment required" clarifies cost. But there is no explicit when-to-use vs. when-not, and no named alternatives, leaving routing among the ~30 siblings to inference.

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