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

Annotations only declare readOnlyHint=false and destructiveHint=false. The description adds genuinely useful traits beyond that: the clustering is deterministic, candidates are verified on their own checks, and the operation incurs no training cost. It omits auth/permission requirements (e.g., the superadmin constraint implied by customerId) and output/return behavior.

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 verb+resource, then supporting clauses; the closing 'Trains nothing and spends nothing' earns its place as a cost/safety signal. Dense but not padded, though the middle clause listing is slightly heavy.

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?

For a non-destructive mutation tool with no output schema, the description conveys what it does and key behavioral traits (deterministic, cost-free, verification-based). Missing only return-value expectations and the permission nuances around customerId, which are minor given the 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 100%, so the schema already documents all four parameters (name, agentId, customerId, splitConditionKeys) including defaults and limits. The description adds no parameter-level detail, so the baseline 3 applies.

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 ('Curate a dataset from an agent's completed runs') and details the internal process (extracts candidates, clusters by type × instruction version × conditions, admits verified successes). This clearly separates it from read/version siblings like get_dataset, list_datasets, and freeze_dataset_version.

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 establishes clear context ('completed runs') and disambiguates scope by declaring 'Trains nothing and spends nothing,' steering the agent away from fine-tuning/optimization siblings. It stops short of naming explicit alternatives or when-not conditions, so it doesn't reach the 5 tier.

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