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

Beyond the annotations, the description discloses that content is frozen and later edits do not alter history, that existing pending requests are reused, and that cloud recipients trigger a human relay with a daily 429 limit and relay/relayNote in the response. It also notes the assistant herself does not receive requests. This is substantial behavioral context not available from the 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?

The most important invocation condition is front-loaded, and every subsequent clause—request-type restrictions, freeze behavior, reuse, cloud relay, rate limit—is an operational necessity. The text is dense but not padded; all sentences earn their place.

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?

Given no output schema and a complex relay flow, the description covers the major behavioral paths: normal send, reuse, cloud-relay with conversation routing, daily limit, and assistant exclusion. It leaves some ambiguity around how GENERAL/TARGETED/LEGACY map to parameters and what note contains, but the core invocation and side-effect profile is complete enough for correct use.

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?

The schema covers only expectedUpdatedAt; the description compensates partially by explaining peerUserId for cloud recipients, conversationId as the assistant DM, and the returned relay fields. However, planId and note are left mostly to inference, and the request-type vocabulary (GENERAL/TARGETED/LEGACY) is not mapped to any input field. With 20% schema coverage, this is adequate but not complete.

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 opens with a precise trigger—only call when the user explicitly asks to send a cooperation request to a specified recipient—and states the concrete action: send the user's ready plan as a private-message card. It also names the request variants (GENERAL/TARGETED/LEGACY), which distinguishes it from nearby send/respond/interest tools. This is a specific verb+resource definition, not a tautology.

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 explicitly gates invocation ('仅在用户明确要求...时调用') and specifies which request types are permitted, with LEGACY requiring prior confirmation of purpose. It also explains the cloud-recipient fallback and directs progress to the assistant conversation, so an agent knows when and how to proceed. There are no sibling alternatives named, but the restriction is unusually concrete.

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