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

Annotations cover the safety profile (readOnly, non-destructive, closed-world), and the description adds substantial context beyond them: the output is a typed tree plus an extensions manifest, screening must be followed by full engine computation with a proof graph on the same snapshot, the binding is a pinnable object, and sourceSchemaContentHash is compared against the pin. These are non-obvious contract details an agent could not derive from annotations.

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?

Purpose is front-loaded in the first sentence, followed by scope exclusions and contract details. Every sentence carries information, but the middle is thick with domain jargon and could be split for readability.

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 an output schema, annotations, and full schema coverage already present, the description still covers the result shape (typed tree + manifest), the two-stage execution contract with snapshot consistency, pin comparison semantics, where provenance facts belong, and what consumes the confidence value. Nothing an agent needs to invoke it 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 baseline is 3, and the description adds real semantics on top: the binding is 'passed as a pinnable object' and sourceSchemaContentHash is the actual hash 'compared loudly with the pin.' The legalTime fallback behavior is already in the schema, so it earns only modest additional credit.

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 verb and resource: 'Build a necessary condition — a superset — for the preliminary selection of rows from an external source.' It also fences itself off from siblings by declaring it is 'NOT a legal answer and NOT SQL' and routes provenance facts to law_ask. The jargon ('manifest of the extensions up to TRUE') is dense, so it is clear but not effortless.

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?

Gives explicit when-not guidance (not a legal answer, not SQL) and names the alternative for adjacent work ('Case facts that carry provenance ... are supplied to law_ask, not here'). It also states the downstream obligation that passing rows must be recomputed by the engine. It never states the positive trigger condition ('use this when you need to pre-filter candidates'), leaving that to inference.

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