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

Annotations supply only idempotentHint=true, so the description carries the behavioral burden and does so thoroughly: it discloses the cascading side effect (hiding all historical and pinned messages in this session plus a broadcast), the blast radius (no account ban, other sessions unaffected), and the repeat-call semantics (重复保持禁言), which is consistent with the idempotent annotation. It even documents what the returned counts mean (hidden = 实际总数, notified = 广播项数, possibly capped at 200, not delivery count) — context that no structured field provides.

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?

The prerequisites and scope limits are front-loaded before the side effects and the return-field caveats, and essentially every clause carries information an agent needs (identity check, side effects, scope, batch prohibition, retry policy, count semantics). It is a dense run-on block with semicolon-chained clauses, which costs some readability but wastes little content.

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?

For a destructive moderation tool with no output schema and minimal annotations, the description covers everything material: authorization, pre-call verification, exact side effects and their scope, non-effects, batching limits, idempotent behavior on repeat, and the meaning of the returned counters. Nothing an agent needs in order to call it safely is missing.

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 description coverage is 100%, so the schema already documents all three parameters, including the instruction to verify the target's nickname/id and to obtain liveId from get_activity or get_organizer_live. The description's 确认活动/场次/范围 and identity-check phrasing largely restates that, adding only mild operational emphasis rather than new parameter meaning, so the baseline 3 applies.

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 names a specific verb and resource in the very first clause (本场禁言 + 隐藏此人本场全部历史发言和置顶), and explicitly bounds it against adjacent actions: 不封号 (not an account ban, distinguishing it from block_user) and 不影响其他场 (session-scoped only, unlike account-level tools). An agent can tell it apart from unban_live_user and block_user without opening any 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?

It states the required caller identity (登录/本场主办方), the pre-call verification step (先主持台核对targetUserId与本人，确认活动/场次/范围), and an explicit prohibition (勿擅自批量). It also gives the failure-handling rule: 未知结果先核验，禁止自动重试, which is exactly the kind of when-to-use/when-not guidance an agent needs for a moderation action.

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