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

Genuinely useful disclosure: the operation does not fix the underlying cause, errored products will fail again on the next publish, selectAll requires confirmCount, and bearer auth is needed. However, 'clears the error records' describes irreversible bulk removal of records while annotations declare destructiveHint=false, and the description never says whether the dismissed errors are recoverable/auditable — a conflict that matters for a 16,000-row bulk operation.

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 tight sentences, front-loaded with the action and scope, then caveats. The '16,000 at a time after a bad import' detail is colorful but earns its place by conveying scale and the manual-work alternative; no filler sentences.

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 5-parameter mutation tool with no output schema, the description covers purpose, the key precondition, and the root-cause caveat but omits what a run returns (e.g., count dismissed, dryRun report shape), what the filters object accepts, and how selectAll interacts with productIds. Adequate but with clear gaps.

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 40%, so the description has to compensate and it only partly does: it explains the selectAll/confirmCount coupling and that the run is refused on count mismatch (also in the schema). It says nothing about the filters object, which is an unconstrained nested object with no schema description, nor about the productIds cap of 5000.

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 (dismiss), a specific resource (listing-error rows), and scope (selected catalog products, in bulk). It is unmistakably distinct from siblings like get-product-listings, get-import-csv-errors, or retry-import-rows; an agent knows exactly what class of operation this is.

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 'otherwise gets cleared by hand, 16,000 at a time after a bad import' framing implies the bulk-cleanup situation where this tool applies, and the caveat that it doesn't fix root causes hints at when it is the wrong choice. But no sibling is named as the alternative (retry-import-rows, fixing the source import) and no precondition beyond confirmCount is stated.

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