Generate an AI Short draft

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

The description goes well beyond the annotations by explaining the full backend pipeline: grounded script, ElevenLabs voiceover, ShortSpec, object storage, and return values. It also discloses the non-publishing behavior, human-approval requirement, and the draft capability that marks this as safe for draft-only tokens. This adds critical behavioral context about side effects and permissions that annotations alone don't convey.

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 every section earns its place: a prerequisite, the tool's outcome, source-kind rules, backend behavior, non-publishing note, and a capability classification. It is front-loaded with the most critical instruction (call prompt_suggest first) and structured in clear segments. No redundant fluff.

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 the tool's complexity (nested source object, multiple source types, multi-stage backend pipeline), the description covers all necessary aspects: inputs, outputs, behavioral constraints, prerequisites, and safety profile. Since there is no output schema, it specifies the returned fields (draftId/spec/script) explicitly. It also provides the capability tag for access control, making it complete for an agent to select and invoke safely.

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 description compensates for the schema's lack of source-kind enumeration by explicitly listing valid values ('autoseo', 'ship-digest', 'trend', 'template', 'raw') and the required nested fields for each. This is essential for the agent to correctly construct the source object, which is the sole required parameter. The schema already describes other params well, and the description adds the missing source semantics.

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 clearly identifies the tool's function: generating a PENDING AI Short draft from an existing source for human review. The verb 'Generate', the specific resource 'approval-ready AI Short draft', the PENDING status, and the explicit 'It never publishes' distinguish it from publishing and other creation tools. The description also names the output artifacts (script, voiceover, ShortSpec, draftId) giving full clarity.

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 instructs to call prompt_suggest first and base the prompt on the top result, providing a concrete usage prerequisite. It states the tool is Pro-only and allowance-capped, and that the result is a draft that cannot reach an audience, implying it should not be used for direct publishing. It doesn't explicitly name alternative tools, but the 'never publishes' and human-review-rail details provide clear when-to-use guidance.

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