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

Annotations already cover safety (readOnly, idempotent, non-destructive, openWorld), so the description's added value is auth/paywall requirements and the env-var cohort fallback, which are genuinely non-obvious. It also discloses the return shape ('clusters and the strongest pairs') and an 'analytics only' disclaimer, though it says nothing about rate limits or cost magnitude per call.

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?

Roughly three sentences with no filler, and every clause carries information (paywall, method, cohort source, output, disclaimer). The premium/API-key bracket is front-loaded ahead of the actual purpose, slightly delaying what the tool does, but it is legitimately actionable.

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 detailed return structure isn't required, and the description still summarizes it (clusters + strongest pairs). With annotations covering safety and the auth path covered in-text, the only outstanding gap is the undocumented topN parameter and the 'at most 20 wallets' constraint interaction with cohort.

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 50%: dex and threshold are documented inline, while topN and cohort are not. The description compensates for cohort (arg or HL_WHALE_ADDRESSES default), but topN — which appears to control how many wallets are scanned — is explained nowhere, leaving a meaningful gap for a filtering knob.

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 and resource: 'Find wallets in a cohort that hold near-identical directional exposure right now (high cosine similarity of their position vectors)' and names the interpretation (same entity, copy-bot, coordinated group). That operation is clearly distinct from sibling lookups like hl_whale_positions or hl_smart_money_score.

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?

Gives a concrete prerequisite path — call hl_request_free_key first for a free key, or pay per call via x402 — and explains cohort sourcing (arg or HL_WHALE_ADDRESSES env var). It does not say when to prefer this over siblings like hl_whale_positions or hl_smart_money_score, so no explicit alternatives are offered.

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