upload_knowledge_file

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

With no annotations, the description carries the full burden, and it delivers meaningful behavior: rate-limit semantics, idempotency expectations, credential sensitivity, async job states (pending/processing ≠ complete), and the metadata-only constraint that forbids local paths and base64. It weakens itself by referencing client_request_id and operation_id, which do not exist in the input schema, creating ambiguity about the actual call contract.

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 text is a run of semicolon-separated fragments with no front-loaded statement of what the tool does, opening instead with idempotency advice. Warnings and workflow rules are interleaved without structure, forcing the agent to parse operational caveats before understanding the core action.

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 complex asynchronous, six-parameter, metadata-registration tool with no annotations or output schema, the description does cover result handling (poll list_topic_directories) and failure modes. It still leaves the parameters and the prerequisite token step unexplained, so an agent could not reliably construct the request.

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 0% across six required parameters, so the description must compensate. It only clarifies the nature of url (remote OSS object, not a local path or base64) and implies topic_id/directory_id through the polling instruction; md5, file_name, and file_type receive no explanation at all.

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 makes clear that this tool registers an already-uploaded OSS object into the knowledge base, accepting metadata only ('OSS 直传成功后将原文件加入知识库，只接收元数据，不接收本地路径或 base64'). This distinguishes it from an actual upload step, which is valuable. It is fragmented and never states the primary action up front, but an agent can derive the verb and resource.

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 gives explicit when-not guidance ('保存遇限流不要重新上传'), a recovery rule (only query the original operation_id, never request a new key), and names the alternative for status checking (list_topic_directories until SUCCESS or FAIL). However, the preceding token-acquisition step (get_knowledge_file_upload_token / get_upload_token) is not mentioned, leaving the overall call sequence incomplete.

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