Evaluate a finite parameter grid against independent human labels: fit and select on train/validation, report test once, and return the candidate without adopting it.
Evaluate an explicit finite parameter grid against independent human labels. Fit/select on train/validation, report test once, return candidate without adoption.
| Name | Required | Description | Default |
|---|---|---|---|
| dataset | Yes | ||
| options | Yes |
Changes observed during successful MCP inspections.
v0.2.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint/idempotentHint/non-destructive, so safety is covered. The description adds real behavioral context beyond that: the train/validation fit-select discipline, the single test report to avoid leakage, and the explicit 'return candidate without adoption' contract, which clarifies that nothing is mutated despite being a tuning run.
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?
Two tightly packed sentences, front-loaded with the core action and then the workflow guarantees. No filler; every clause (grid, labels, split discipline, non-adoption) earns its place.
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?
The workflow prose is reasonably complete for a train/val/test tuning tool, and no output schema exists to explain returns. But with two fully undocumented nested-object parameters and no detail on the option schema or result shape, the definition is thin for a fairly complex grid-search operation.
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% and both parameters are undocumented nested objects. The description gestures at them ('explicit finite parameter grid' -> options, 'independent human labels' -> dataset) but never explains their expected structure or required keys, leaving the agent to guess the grid format.
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 gives a specific verb+resource: it evaluates a finite parameter grid against human labels and reports test performance. It is clearly a tuning/selection operation distinct from cataloging or applying calibrations. It does not, however, name any sibling (e.g., ap_evaluate or ap_fit_calibration) to sharpen the boundary.
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?
Usage is only implied. The phrase 'return candidate without adoption' hints that a separate adoption step exists (e.g., ap_apply_calibration), but the description never states explicitly when to prefer this tool over ap_evaluate or ap_compare. An agent must infer the routing.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.