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

Annotations already cover read-only, idempotent, non-destructive, but the description adds material context an agent could not infer: transitive match grouping, the keep/merge policy semantics, the returned payload shape, a hard performance envelope (5,000 records stay fast), and the paid x402/Solana pricing plus an alternate HTTPS endpoint. Cost and payment requirements are exactly the kind of behavior annotations never carry.

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?

Purpose and domain come first, then modes, then output behavior, then commercial terms — a sensible front-loaded order with no filler. The single dense parenthetical block listing normalization rules is information-rich but heavy to parse; slight trimming would help.

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 8 parameters, a nested types object, and no output schema, the description still covers mode semantics, retention policy, merge behavior, and the return shape (unique records plus duplicate groups with scores). Payment and alternate-transport details are included, leaving nothing an agent needs in order to call it 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?

Schema coverage is 75%, but the two undocumented enum params (mode, keep) are fully explained in prose, including the normalization sub-rules per field type and the three keep strategies. It adds real meaning beyond the schema (threshold is described only via 'weighted similarity with weighted fields' rather than tied explicitly to the 0.5-1.0 range), while types/fields/includeRecords rely on their own schema descriptions.

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?

States a specific verb pair (find and remove) and resource (duplicate records) scoped to a JSON array, with concrete domain examples (CRM contacts, leads, customers, products). It also enumerates the three operating modes, so an agent can distinguish it from generic siblings like fuzzy_match or text_similarity without opening any schema.

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?

It explains what each mode does (exact, normalized with per-type normalization rules, fuzzy with Jaro-Winkler/token similarity and blocking), which implicitly guides mode selection, and the schema supplies defaults. What is missing is explicit routing language: no statement of when to prefer this over fuzzy_match or text_similarity, and no explicit failure/exclusion conditions.

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