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

Annotations say readOnlyHint=false but the description explains exactly what is written ('saves only people the role has not seen') and that the call is free, unscored, and resumable from the last round. The request_id retry clause reconciles the idempotentHint=false annotation by explaining that idempotency is opt-in via request_id rather than default. This is materially more than the annotations convey.

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?

Front-loaded with the core action, then the exhaustion path, then cost/retry caveats — a sensible ordering. The prose is a single dense run-on with stacked parentheticals, which costs some readability, though nearly every clause carries unique information.

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 covers the key return signals (round.changes, round.exhausted, the broaden suggestions) and the empty-result semantics, which is the critical part for an agent. It stops short of describing the shape of the returned people list, but that gap is minor relative to what it does disclose.

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%, with only request_id documented in the schema; the description reinforces its retry semantics and adds the intent behind it. role_id is only implied by 'a role's plan' and never explicitly defined, which is the one gap.

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 ('Run the next round of a role's plan') and immediately disambiguates it from the search/plan siblings by describing the resumable, plan-driven behavior. An agent can tell this apart from find_people_for_role, search_people, and set_search_plan without opening a schema.

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?

Explicit preconditions ('Needs a plan (create_role makes one)'), explicit routing for the terminal case ('suggest a wider search from broaden ... approve one with set_search_plan'), and explicit follow-up ('call score_role_person for the people worth it'). It also states the retry protocol with request_id, leaving nothing to inference.

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