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

The readOnlyHint annotation covers safety, and the description adds substantial behavioral context beyond that: results can include matches inside videos with an exact video_moment timestamp, and each image_url is only a 600px preview rather than the full original. This materially changes how an agent should use the returned data.

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 dense but well structured: core capability first, then mode definitions, then scoping/filtering, then special result behaviors, then a caveat with an alternative tool. Every sentence adds operational value, and the length is justified by the tool's 15 parameters and lack of an output schema.

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 complex search tool with no output schema, the description covers the critical gaps: mode interactions, required versus optional inputs, result shapes like video_moment and regions, preview limitations, and the pointer to list_generations for full-fidelity originals. Remaining details like limit and offset are already present in 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 high at 87%, so the description does not need to re-document parameters. It adds value by explaining mode-dependent semantics such as query being required for text mode, similar modes requiring anchors, and include_regions producing bounding-box data, plus the video_moment consequence of certain text results.

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 verb and resource: detection-aware search across the user's assets and generations. It distinguishes itself from title/prompt-based retrieval and from sibling listing tools by emphasizing what the scanner SAW inside each image, and it breaks out four distinct modes so an agent immediately understands the tool's scope.

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 mode-specific usage rules: text requires a query, color ranks by hex closeness, similar needs an anchor, and browse requires no query and is for exploration. It also instructs when to step outside this tool, telling the agent to re-read the original from list_generations before using a hit in a document.

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