Find counterparties

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

Annotations already declare readOnlyHint, idempotentHint, openWorldHint, and non-destructive, so the safety profile is covered. The description adds the meaningful behavioral fact that results are computed over a bounded "transfer sample" rather than the full history, which is a real caveat an agent should know. It does not address pagination or result shape, but the sampling disclosure is substantive added context.

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?

Two sentences with zero filler: the first front-loads what is computed and how it is broken down, the second gives the motivating use case. Every clause earns its place.

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 and 17% parameter coverage, the description does well on core behavior and sampling but leaves gaps: how "top" limits results, what sinceUnix filters, and whether the sample bound makes output approximate in a way the agent should caveat. Adequate but not fully complete for a 6-parameter analytic tool.

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 only 17% (just the address field), so the description must carry the load and it partially does: "transfer sample" maps to transferSample, "inbound/outbound" to direction, and "per-mint volume" to mint. It says nothing about top, sinceUnix, or the ranking/limit semantics, leaving real gaps.

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 states a specific verb and resource with unusual precision: it ranks every address a wallet transacted with, split by inbound/outbound count and per-mint volume, plus first/last contact. An agent can tell what it produces. However, it never names a sibling (e.g., analyze_wallet, trace_funds, find_cluster), so the boundary against adjacent tools is left to inference.

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 to spot the funder, the sink, or the one address a wallet keeps cycling with" gives intent-level guidance that implies the analytic scenarios this tool serves. It offers no explicit when-not condition and no named alternative, so an agent choosing between this and trace_funds or analyze_wallet gets only implied direction.

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