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

The annotations already declare readOnlyHint=false, openWorldHint=true, idempotentHint=false and destructiveHint=false, so the description's added value is what happens downstream: the report enters 'private review' and is not a status claim. It also states an authorization requirement ('your own operating authority') and hard redaction rules, which are real behavioral constraints. It does not mention rate limits, review latency, or the consent_confirmed and idempotency_key mechanics present in the schema.

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 short sentences, with the core action stated first, followed by disposition and then constraints; there is no filler. 'Under your own operating authority and platform rules' is the only slightly oblique phrase, but overall the description is tight.

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 output schema means return values need not be described, and the annotations carry the safety profile. However, for a 15-field evidence-submission tool, key semantics such as what distinguishes 'passed' from 'abandoned', when error_class is meaningful, and what consent_confirmed means are all missing.

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?

The single parameter is a nested object with roughly fifteen fields, and its schema_description_coverage is 0%: outcome, error_class, attempt_env, challenge_kind, challenge_vendor and consent_confirmed are all undocumented. The description does help in one concrete way: the 'never include names, account details, URLs, cookies, tokens' rule tells the agent that site_host must be a bare host and that payloads must be scrubbed. That is nowhere near enough to compensate for a fully undescribed field set.

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 states a specific verb and resource: 'Submit one de-identified firsthand observation.' It also delimits the scope ('public reports enter private review, not site-status claims'). It never mentions the sibling check_incident, so an agent gets no explicit differentiation between reporting and checking an incident.

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 'under your own operating authority and platform rules' only implies that this is the write path for first-hand evidence, and 'are not site-status claims' implies a boundary but names no alternative tool. When to use this instead of check_incident is left entirely to inference.

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