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

With no annotations at all, the description carries the full burden and does unusually well: it pre-announces that some endpoints return the literal status 'not_served' with a reason, explains the validation rationale (no usable ranking skill on held-out EPA ECOTOX values, larger error than a fixed baseline), and warns that no GHS aquatic category is produced and the PBT T criterion is 'unassessed' rather than negative. It never mentions the api_key credential requirement or any rate/credit limits, which is the remaining behavioral gap.

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 capability before the caveat, which is the right order, and every sentence carries routing or expectation-setting value. The middle rationale ('no usable ranking skill and a larger error than a single fixed number') is slightly more argumentative detail than an agent needs to select the tool, but it justifies the 'not_served' behavior that would otherwise look like a bug.

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-value documentation is not owed, yet the description still explains the most confusing return semantics (not_served placeholders, absent GHS category, unassessed PBT T criterion), which is the right use of the space. The credential parameter is the one element left underexplained for a tool that has zero annotation coverage.

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 largely does not: it never mentions 'smiles' (only implied by 'structural alerts') and never mentions 'api_key' at all, including whether it is required, where to obtain it, or whether an empty default works. An agent must infer input format and credential handling entirely from bare property names.

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 opening clause names the verb+resource precisely: an environmental fate screen covering BCF, biodegradation half-life/ready biodegradability, ecotox structural alerts, and REACH P/B criteria. It then explicitly states what it is not (acute aquatic toxicity), so an agent can distinguish it from any acute-tox or generic tox sibling without opening the schema.

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?

Explicit when-not guidance: no acute fish/Daphnia/algae endpoints, and a named alternative route (measured OECD 203/202/201 or EPA ECOTOX Knowledgebase) for those endpoints. The exclusion is front-and-center rather than buried, which is exactly what routing decisions need.

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