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

Annotations already declare readOnlyHint, idempotentHint and destructiveHint=false, so safety is covered. The description adds real behavioral value beyond that: the records carry a key, the reason given and a timestamp, and a human-readable summary is exposed in the `say` field. What's missing is pagination/volume behavior and ordering of the returned list.

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, front-loaded with the identity of the resource, then the returned fields, then the recovery action and summary field. No filler, though the opening clause is awkwardly worded and costs a small amount of parsing effort.

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 no output schema, the description must carry the return-value burden, and it does so at a high level (key, reason, when, `say` summary). For a no-parameter read tool this is nearly sufficient; it stops short of describing ordering, limits, or what happens when nothing has been dismissed.

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 tool takes zero parameters, so the baseline is 4 and there is nothing for the description to clarify on the input side. The description instead spends its words characterizing the returned record shape, which is appropriate.

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 uses the verb 'Matches' plus a specific resource scope: dismissed people and whole companies that were ruled out. An agent can tell this is a listing of rejections rather than active matches, though the phrase 'the person dismissed' is slightly ambiguous (it reads as a status of a match, not an action on a person). It never explicitly contrasts with best_matches, so differentiation from siblings is only implicit.

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 only routing guidance is 'Use restore_match to put one back,' which directs the agent to a sibling for the follow-up action rather than explaining when to call list_dismissed instead of best_matches or list_actions. The context of use (reviewing what was ruled out) is implied rather than stated, and no exclusions are given.

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