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

Annotations supply only destructiveHint=false, so the description carries the real behavioral load and does so well: card content is frozen so later plan edits do not alter history, pending-response requests are reused, cloud-user sends are relayed by a human secretary via phone/WeChat with progress tracked in the secretary conversation, and the hard cap of 5 relays per person per day returns 429 cloud_relay_daily_limit. Rate limits and immutability are exactly the context annotations do not provide.

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?

Appropriately short and front-loaded with the use condition before the behavioral detail. It loses a point for dense semicolon-chained clauses and for naming identifiers that do not appear in the schema, which undercuts scanning.

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 5-parameter send operation with no output schema, the description covers consequences and limits well but leaves 4 of 5 parameters semantically unexplained and introduces phantom field names. An agent can decide whether to call it but may still be unsure what to put in planId versus note, or how a pending reuse affects arguments.

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 only 20% (only expectedUpdatedAt is documented), so the description must compensate and only partly does: it clarifies that peerUserId must still be passed for cloud users and that conversationId is the caller's own chat with the secretary. planId and note are never explained, and worse, the text references fields that do not exist in the schema (targetUserId, relay, relayNote, isCloud), which can send an agent looking for parameters it cannot supply.

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 concrete verb and resource: sending one's own ready cooperation plan as a private-message card ('发送本人就绪方案私信卡'). That is specific enough to distinguish it from respond_cooperation_request or send_message, though the sentence is larded with jargon (GENERAL/TARGETED/LEGACY plan types, 小秘书 relay) that an unfamiliar agent must decode.

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 gating is given: only call when the user explicitly asks to send to a named recipient, only GENERAL or TARGETED plans whose targetUserId equals the recipient, and never send to the human secretary. What is missing is sibling routing — nothing says when to prefer send_cooperation_interest, send_message, or request_activity_cloud_relay instead.

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