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

Annotations already establish a read-only, idempotent, non-destructive, open-world profile; the description adds materially more: no model involved, nothing stored, a signed record produced, a 500-row cap, no pooled data (table built from the caller's rows only), and an explicit external eCFR lookup triggered by check_sources. Those are exactly the data-handling and side-effect facts an agent cannot get from the annotations.

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 lead sentence front-loads purpose, then input spec, then output spec, so a reader can stop early. It is one long dash-chained block, however, and the density makes scanning harder than a few separated sentences with the row cap and privacy rule split out would be.

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?

There is no output schema, so the description carries the full return-value burden and does it: it enumerates every output (win rate with 95% interval, median/90th percentile/mean days to close, three groupings, code meanings, small-group flag, Reg E 1005.11 and Reg Z 1026.13 clocks, signed record). Limits, privacy constraints, and the external-check toggle are all covered.

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?

With 67% schema coverage the baseline is 3, and the description earns above that by explaining the semantics of min_cases ('a flag for groups with too few cases') and check_sources ('ask the eCFR whether either rule was amended since'). It also reiterates the network and reason-code formats without adding syntax beyond the schema.

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 opens with a concrete verb+resource ('Card dispute outcome table') and then specifies inputs ('one row per card dispute case') and exact outputs ('win rate ... and days to close ... by network reason code'). It never names or routes against siblings such as outcome_dispute, dispute_record, or outcome_case, so an agent must infer the boundary itself.

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?

Usage is implied by the instruction to submit one row per handled case, plus data-entry rules ('Use merchant labels, never card numbers, names or emails', 'Up to 500 rows') and a scope disclaimer. There is no explicit when-to-use/when-not guidance and no signposting of an alternative tool for related tasks like recording a single dispute.

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