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

Annotations only declare idempotentHint/destructiveHint, and the description adds context the agent cannot infer: the 【需要登录】 auth prerequisite, the visibility scope ('我能看到的消息圈'), what the result set omits (comments and likers), and a real cost warning that the semantic leg triggers a paid vector call. That is exactly the behavioral load annotations do not carry.

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 four dense sentences with the auth flag bracketed up front, then scope, then examples, then the follow-up chain, then the cost caveat. Nothing is padded and the most decision-critical constraint (paid vector call) is placed last as a warning rather than buried.

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 two-parameter, no-output-schema search tool, the definition covers auth, visibility scope, matched fields, result omissions, next-step tools, and cost. An agent can decide whether to call it, how to phrase q, and what it will and will not get back.

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 q and limit are documented structurally (including the 1-30 bound and default 10). The description still adds meaning by clarifying that search is hybrid keyword-plus-semantic over specific fields, which tells the agent how to phrase q (keyword or a sentence) rather than guessing at matching behavior.

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 states a specific verb and resource ('按关键词 + 语义搜我能看到的消息圈') and enumerates the searched fields (正文、作者昵称、挂的卡片标题), so an agent knows precisely what is matched. It does not explicitly contrast itself with the many feed-listing siblings (list_moments_feed, list_ecosystem_moments, personalized_feed, search_all), so sibling differentiation is implied rather than stated.

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?

Two concrete example queries ('最近谁在说宠物硬件打样', '有没有人发过出海渠道的资源') make the intended use unambiguous, and the arrow chain 本工具 → get_moment → get_creator → start_conversation tells the agent what to do next. It also gives an explicit operational exclusion ('别循环调' for the cost-bearing semantic leg), though it never says when to prefer a plain feed-listing sibling instead.

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