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

Annotations only declare destructiveHint=false, so the description carries the behavioral load and does so: it states login is required, that a billing model is invoked, that looping retries are wrong, and that 429 responses require backing off. It further discloses merge/dedupe semantics and that the top score plus all reasons are preserved, which is real behavior an agent could not infer from 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?

Every clause carries information — cost model, rate limits, merge behavior, parameter strategy, workflow position, and the bUserId fallback — with little filler. The cost is structure: semicolon-chained clauses front-load warnings instead of the tool's purpose, making it dense to parse on first read.

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, the description still exposes the shape of results (score, reasons, leadId possibly null, bUserId) and the auth and billing preconditions. It lacks detail on result volume per anchor or how reasons are ordered, but covers enough for correct invocation and downstream handoff.

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 100%, so the schema already defines both parameters and their bounds, making 3 the baseline. The description adds strategy beyond the schema: multiple leadIds for broad pool scanning versus a small perLead, and perLead=20 as the escape hatch for a difficult single anchor.

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 conveys a specific action and scope: one full-site search per anchor lead, merging private-pool and site-wide candidates with cross-anchor dedupe and highest-score retention. It is distinguishable from siblings (list_broker_leads feeds it, create_broker_match consumes it). However, the purpose is buried behind operational warnings rather than stated up front, so an agent must read several clauses before knowing what the tool returns.

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?

The pipeline 'list_broker_leads → 本工具 → 念候选给用户确认 → create_broker_match' explicitly names when to use it and which siblings precede and follow it. It also gives selection heuristics: many leadIds with small perLead for pool-wide scanning, perLead=20 for a hard single match, plus a fallback instruction when candidate leadId is null.

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