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

Annotations declare readOnlyHint, idempotentHint and openWorldHint=false, so safety is already covered and the description is free to spend its words elsewhere – which it does. It discloses behavior the annotations cannot contain: matching ignores case and accents, and when nothing matches, a `closest` field returns near spellings, with a worked example ('vip' → 'VIP Clienti'). It omits any note on volume, ordering, or what an empty workspace returns.

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 sentences, front-loaded with the enumeration and then the decision rule, with the fuzzy-match nuance last. The prose enumeration of the seven kinds is partly redundant with the enum, but it earns its place by telling the agent what each kind actually represents rather than just repeating the token.

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?

An output schema exists, so return structure needn't be explained, yet the description still volunteers the one output behavior an agent would otherwise miss – the `closest` fallback. For a read-only three-parameter lookup this is nearly complete; the only gap is the unresolved relationship to the dedicated reader tools covering the same resources.

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 67%, so the schema already documents `match` and `workspaceId`; the description still adds meaning the schema does not, describing what `match` accepts in human terms and, more importantly, what happens on a miss (`closest` with the nearest spellings). `kind` is left entirely to the enum, which is acceptable since the enum is self-describing, though the prose enumeration adds the '(with ids)' detail for segments.

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?

Names a specific verb and resource – reads the exact stored values a workspace uses for one kind of thing – and enumerates all seven kinds (segments with ids, contact sources/statuses, prompt topics, prompts, competitors, keywords). An agent knows precisely what comes back. It does not, however, distinguish itself from siblings like readSegments, readTopics or readCompetitors, which read some of the same resources, so the boundary remains ambiguous.

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?

"Use it before a filter or before adding a prompt, a keyword or a competitor that may be there already" gives a concrete trigger condition tied to real workflow moments. It stops short of the exclusion case – it never says when to prefer the dedicated readers (readSegments, readTopics) instead, leaving the overlap with siblings unresolved.

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