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

Annotations already indicate readOnlyHint=false, destructiveHint=false, idempotentHint=true, openWorldHint=true. The description adds valuable context: it discloses that the report is not taken on faith (a subsequent read confirms or reopens), and that it requires execute access. It also mentions the 'fix_reported' status. This goes beyond the annotations by explaining the post-condition and verification workflow, which is critical for the agent to understand the asynchronous nature.

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 description is three sentences: first states the action and purpose, second explains the downstream effect, third clarifies the verification and permission. It's well-structured, front-loads the verb 'mark it fix_reported', and every sentence adds distinct meaning. No fluff, efficient use of words.

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?

Given the tool's moderate complexity (2 params, no enums) and the absence of a detailed output schema (there is one but not shown, the description doesn't need to explain return values). The description explains the workflow, the expected effect, and the verification process experimental. It covers the essential points: when to use, what happens after, and that it requires execute access. The only minor missing piece is explicit parameter names, but that's covered by the schema itself (which has names and types). So it's complete enough for an agent to call correctly.

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 schema has 0% description coverage, so the description must compensate. The description mentions 'findingId' implicitly but does not explain the 'note' parameter at all. It says 'with a short note' which maps to the 'note' parameter, but doesn't specify max length or optionality. It could have explained that 'findingId' is the UUID of the finding to mark, and 'note' is optional text. Given that there are only 2 params usable via description, the coverage is slightly better than low, but it's still not explicit enough for an agent to know exactly what to pass. However, because the note is described, it gets a 4.

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 clearly states the action ('mark it fix_reported'), the resource ('a finding'), and the context (after fixing meta description, robots rule, or llms.txt). It distinguishes itself from siblings like 'report_publication' which is about publication, not findings. The phrase 'Report a finding as fixed' is not a pure tautology because it clarifies which domain (finding) and action (fixing).

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 tells when to use it: 'After the agent fixes a finding on the site'. It explains the post-condition (shows as done on the customer's Fixes list) and that the system will later confirm or reopen. It doesn't explicitly name alternative tools or when not to use it, but the context implies it's the right tool for marking a fix; 'Requires execute access' hints that not all agents will have this tool, but doesn't list alternatives. Could have explicitly said 'Use this only after fixing a finding, not for reporting publications (use report_publication) '.

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