从课件生成讲解并按门禁写入

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

With no annotations, the description carries the full burden and does well: it discloses that nothing is written by default, that failed audits do not get beautified, that apply rolls back on broken formulas, and that no model is invoked. Missing are details on permissions, rate limits, and what the audit failure reason looks like, but the write/rollback semantics are unusually clear for an unannotated mutation path.

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?

Front-loaded with the skeleton return and the draft-input rule, then escalates through review and apply gating in the order an agent would use them. Dense but every clause carries a rule; it is slightly run-on with several short declarative fragments that could be grouped more tightly.

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 7-parameter, multi-branch tool with no output schema, the description documents the control flow thoroughly but says nothing about return values — what the skeleton contains or what an audit result looks like — leaving the agent unable to plan around the response. The flow is adequately covered; the response shape is not, which is the remaining material gap.

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, but the description adds cross-parameter meaning the schema cannot: it ties review=true to the parentPageId write target and ties apply=true to a prerequisite content-audit pass. That interaction is the part most likely to be misused and is spelled out explicitly. Only themeId and title get no discussion in the description, which is acceptable since the schema covers them.

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 opening line states a concrete verb and resource (read the deck and return a structural skeleton), then clarifies the tool's actual job: assembling a lecture draft through audit gating and optional write. It does not name sibling tools like draft_lecture or ingest_deck, so the agent cannot distinguish it by name alone, but the multi-mode pipeline behavior is unmistakable.

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?

The description gives explicit mode-selection rules: omit draft for skeleton only, pass draft to run content audit, set review to write the pending draft, set apply to beautify only after passing. That is precise when/when-not guidance for the tool's own branches. The gap is that it never contrasts itself with sibling tools such as draft_lecture or preview_page, so the agent must infer which pipeline entry point to use.

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