Generate video (Veo, like Google Flow)

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

With no annotations, the description carries the full burden and does so: 1-6 minute render latency, default blocking behavior via wait, the pending/operationName fallback path, and that files are written locally with their paths returned. It also flags that image arguments accept either asset ids or local file paths, which is non-obvious side-effect-adjacent context.

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?

Purpose is front-loaded and every sentence adds information; the prompt-writing and latency sentences both earn their place. It is a single dense block rather than structured sections, so the mode list and the async-fallback instruction take a careful read to separate.

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?

Covers the things an agent cannot infer: render time, blocking vs non-blocking, where output files land, and the pending-operation handoff. For a 16-parameter, no-annotation, no-output-schema tool it omits cost implications and safety-relevant behavior (personGeneration, content restrictions), which the agent must infer from the schema.

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 already 88%, so the 3 baseline applies, but the description adds genuine value by tying startImage/endImage/referenceImages to named modes and by spelling out what a good 'prompt' should contain (subject, action, camera move, lens, lighting, dialogue in quotes, SFX). It leaves several params (numberOfVideos, resolution, aspectRatio, negativePrompt, personGeneration, fast, seed) entirely to 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?

States a specific verb and resource ('Create a short video clip with Veo') and immediately enumerates the three generation modes, which cleanly separates it from flow_generate_image (stills) and flow_extend_video (extending an existing clip). The scope constraint 'short video clip' plus the duration bounds makes the intent unambiguous.

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?

Explicitly maps each mode to its required inputs ('text-to-video (prompt only)', 'frames-to-video (startImage, optional endImage)', 'ingredients-to-video (referenceImages...)'), states the prompt-unless-startImage condition, and routes the agent to flow_check_operation when status=pending. It even points at flow_list_assets / flow_generate_image as the source of image ids.

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