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

Annotations already declare read-only, idempotent, non-destructive, closed-world behavior, so the bar is lower. The description still adds context the annotations do not: a null request silently triggers the shipped synthetic example, and supplied data requires an explicit rights flag. The caveat 'this does not verify a licence' honestly discloses that the flag is an attestation rather than a check.

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 short sentences, no filler, with the capability statement front-loaded and the two operational rules following. Every sentence contributes a distinct fact: methodology, default mode, and the licensing gate.

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?

An output schema exists, so return values need no prose. The default-path and rights-gate behavior are covered, which is the highest-risk ambiguity. However, for a tool taking a large nested request with covariance matrices and up to 100k positions, the description says nothing about validation expectations, size/format constraints, or what happens when rights_confirmed is false.

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 carries the full burden, yet it only clarifies the two top-level parameters (request may be omitted; rights_confirmed must be true for supplied data). The large nested ParametricRequest object — nav, as_of, data_mode, covariance matrix shape, confidence bounds, horizon_days, positions structure — gets no semantic guidance beyond the enum values visible in the schema. It adds real meaning for the two entry points but leaves most parameters undocumented in both places.

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 names a specific methodology and output set: parametric normal VaR and ES, Euler component contributions, and incremental VaR per position. That is far more precise than the bare name, and 'parametric normal' implicitly separates it from historical/filtered attribution siblings. It stops short of explicitly naming the sibling it displaces, so the agent must infer the boundary from methodology words alone.

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?

It gives conditional guidance: omit the request to run the shipped synthetic example, and when supplying data require permission and set rights_confirmed=true. That covers two concrete entry paths. It never states when not to use this tool or which sibling handles ES-only, historical, or non-normal variants, so routing between the many risk_* tools is left to inference.

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