generate_storyboards

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

Since no annotations are provided, the description carries the full burden of behavioral disclosure. It clearly states the operation is destructive (replaces the whole episode's storyboards, loses already-generated images, costs money), asynchronous (returns immediately with status:generating), slow (5–10 minutes typical), and gated by a review token. It also discloses token invalidation, default error blocking, and the customer-decision rule for acknowledge_review.

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 description is long but well-organized with ★ markers, bolded warnings, and critical information front-loaded. Each block has a clear purpose, though the repeated waiting-time warnings ('分钟级', '5~10 分钟', '单次 LLM 调用 3~7 分钟', '60 秒…3 分钟内查不到分镜都是正常的') are somewhat redundant and could be condensed without losing the anti-resubmission message.

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 destructive, asynchronous tool with no output schema and no annotations, this definition is unusually complete: it covers async return, polling, progress checking, replacement behavior, review gating, and quote confirmation. It falls short only by not explicitly explaining where a run identifier comes from for get_run_status and by leaving episode_id semantics implicit.

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 high (80%) and the schema already documents most parameters. The description adds extra meaning for review_token (token invalidates if content changed; can re-review for free) and acknowledge_review (don't decide for the customer), and clarifies that quote_id comes from a quote_* tool after user approval. However, episode_id has no schema description and the tool description only implies its role via '整集'.

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 quickly identifies the action ('确认后拆镜' / storyboard generation), the target resource (整集分镜), and the asynchronous execution model ('后台跑,立即返回 status:generating'). It also differentiates itself from sibling tools by telling the agent to poll with get_storyboards and check run progress with get_run_status.

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 required prerequisites and exclusions: do not call before review_script (or you get 400), do not confirm without an approved quote_id, use confirm_replace=true only when knowingly replacing existing storyboards, and do not resubmit while a run is active. It names the relevant sibling tools (review_script, quote_*, get_storyboards, get_run_status) and explains which one to use when.

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