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

Beyond the annotations it adds material the agent could not otherwise know: a strict no-side-effect boundary ('no network request, fetch, money movement, message, sharing, configuration change or external write'), pricing and the free-call allowance, the required X-Wever-Wallet-Proof binding to the exact JSON arguments and route, and the fact that a signed Ed25519 receipt is returned. It also enumerates the four verdict values (headline_only, disclosure_gaps, disclosure_unverified, fully_disclosed), which is genuine behavioral disclosure. This is well past what readOnlyHint/openWorldHint/destructiveHint convey.

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?

The core is one long run-on sentence that enumerates the same field set twice — once as what is submitted, once as what the signed report names — which makes it slow to parse and partially redundant. The second enumeration earns some of its space because it describes the returned findings and per-item fixes in the absence of an output schema. Front-loading is acceptable, but the structure is a wall of clauses rather than scannable segments.

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?

With seven parameters, nested objects, no output schema and only 43% schema coverage, the description still covers input intent, paid/auth mechanics, verdict vocabulary, and what the returned report contains. It also points to the contract endpoint and human page for deeper specification. Nothing an agent needs in order to call this correctly is conspicuously absent.

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?

At 43% schema description coverage the description has to carry weight, and it does: it explains that headline entries need spread/sample-size/baseline flags shown beside each score, that limits need disclosure state, placement and content, that failures need reported-vs-known counts plus examples and placement, and it spells out the run conditions (model, dataset, run date, scoring method, settings) and the full_results tri-state. The enum meanings are restated in prose ('beside the headline, a later section, an appendix or a separate document'). Only the top-level 'label' parameter goes unmentioned, so it lands just below the top band.

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 action and the artifact — 'submit one evaluation summary' — and then enumerates exactly what is examined: headline scores with spread/sample size/baseline, limits with placement, failures with counts and examples, run conditions, and full-results availability. It also names three adjacent checks (wever_ai-disclosure-check, wever_recommendation-disclosure-check, wever_benchmark-evidence-check) and their distinct scopes, so an agent can separate this from disclosure-adjacent siblings. The density of the enumeration makes the core purpose harder to extract than it should be, keeping it out of the top band.

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 phrase 'one evaluation summary written for a buyer or principal' sets the input context, and the sentence starting 'This reads what an evaluation summary discloses beside its headline:' routes three related-but-different questions to named sibling tools. That is real alternative routing rather than implied usage. It stops short of stating when-not-to-use versus the closest siblings (wever_evaluation-result-check, wever_evaluation-reproducibility-check), so a 5 is not warranted.

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