offline_html

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

Annotations already mark destructiveHint=true and readOnlyHint=false, and the description adds substantial context: permanent deletion scope, retention of local registration and non-deletion of local HTML, cache non-clearing, version-control refusal, and recovery from partial cleanup via returned suggestions. No contradiction with annotations; idempotentHint is supported by retry semantics.

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?

Four brief sentences front-load the outcome and then pack safety conditions, side effects, and recovery guidance without redundancy. Every sentence contributes operational value.

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?

The description is complete for a destructive operation: prerequisites, refusal criteria, side effects, and partial-cleanup handling are all present. It loses one point only because there is no output schema and the description references '返回建议' without specifying what a successful return contains, leaving a minor ambiguity for the agent.

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 baseline is 3; the description adds meaning beyond field names by tying expectedSha256 to the get_html result and retry behavior ('不删除已变化的内容') and explaining localPath does not read/delete local HTML. This compensates slightly beyond the schema.

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 specific action and outcome: '下线指定页面：永久删除当前云端 HTML 及该站点旧版快照，保留本地登记', clearly distinguishing this destructive offlining operation from siblings like get_html/list_html/online_html. It states the resource (current cloud HTML and old snapshots) and the retained part (local registration).

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-to-use and workflow conditions: '必须用户明确要求' and '先 get_html 查询哈希', plus refusal conditions ('启用、暂停或无法确认时拒绝删除'). It also clarifies that external caches are not cleared, managing expectations. This is stronger than a generic 'use for offlining' hint.

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