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

Annotations already declare readOnlyHint, idempotentHint, destructiveHint=false and closed-world, so the safety profile is covered and the bar is lower. The description does add real value beyond them: it discloses that the figures are counts only, derived from a sample ('share of sampled answers'), and explicitly warns the output is not a verdict — a meaningful interpretive constraint. It still does not say whether the payload is network-level or per-Agent, nor what shape the response takes.

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, front-loaded with a compact enumeration of what is measured, then the interpretive caveat, then a pointer to the method at /about#sybil. Every clause carries information; the metric list is dense but scannable and no sentence is filler.

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 input parameters and no output schema, the description carries the full burden of explaining what comes back, and it does so at the semantic level by listing the measurement families and flagging the sampling basis. The remaining gap is structural: it never states the granularity of the result (network-wide vs per-Agent) or its format, which an agent needs before wiring the output into a judgment.

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?

The tool takes zero parameters, so there is nothing the description needs to disambiguate; baseline applies. The only semantic constraint worth noting — sampling and count-only reporting — is captured in the purpose text rather than around any input.

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 enumerates the exact signals being reported (slot-template share, answer families, capped payees, inviters, web labels, shared wallets, anchor-linked Agents) and names the reporting entity (POAScan), so an agent knows precisely what this resource contains. It stops short of a clean verb-plus-resource framing such as 'returns network-wide sybil metrics' and never states whether results are per-Agent or network aggregate, which slightly blurs the purpose against a sibling like get_cluster_signals.

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 to judge how much of the network looks scripted or linked' gives a clear selection criterion for this tool. The added caution that these are measurements with innocent explanations and 'not a verdict on any Agent' is an effective when-not-to-overinterpret rule, but no alternative diagnostic tool is named for cases where the agent instead wants per-cluster signals.

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