Start an image search session

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

With no annotations, the description carries the full burden of behavioral disclosure. It details the process steps, the return values (grid file path, selected ids, metadata table, keywords used), and explains special behaviors like the 'filter' parameter (false skips filter) and 'keywords' parameter (skips LLM parsing). It also notes the default for filter follows FILTER_MODE. This is transparent, though it does not mention session state persistence or file system side effects beyond returning a 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?

The description is two concise sentences. The first enumerates the pipeline steps without redundancy, and the second lists the return values. It is front-loaded with the core action and contains no filler or repetitive language.

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?

The description covers the tool's behavior, parameters, and return values. It does not explain the overall session workflow (e.g., that image_search_iterate follows), but that is arguably outside the scope of a single tool description. The return list compensates for the absent output schema. Completeness is high for the tool's complexity.

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 83%, so most parameters are described in the schema. The description adds meaningful context beyond the schema by explaining that 'count' defaults to grid capacity, 'filter=false' skips the vision filter, and 'keywords' skips LLM parsing. It also clarifies the filter default via FILTER_MODE. This enriches the parameter understanding.

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 clear verb and resource: 'start an image search session.' It then describes the full pipeline (parse query, search parallel, dedupe, render grid, filter, return results). The action is specific and distinct from siblings like iterate, collect, and status, making it obvious this tool initiates the session.

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 does not explicitly state when to use this tool versus its siblings. It does not mention 'use this to begin a session' or direct the agent to image_search_iterate for refinement. While the name implies it is the starting point, there is no explicit guidance on choosing it over alternatives or any exclusions.

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