instagram_search_start

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

Annotations declare readOnlyHint, openWorldHint, idempotentHint, and destructiveHint false. The description adds substantial behavioral context: it opens a job on disk that survives crashes and rate limits, runs non-blockingly over many calls, and explains that some attributes are advisory and never affect ranking. This goes far beyond the annotations and gives the agent a clear model of the tool's side effects and limitations. No contradiction with annotations.

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 well-structured with a clear opening, an explanatory paragraph, a list of attributes, and a structured Args/Returns section. Every sentence contributes essential information—usage context, behavioral notes, next steps, and return format. No filler or redundancy, and important information is front-loaded.

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 (multi-stage search, nested persona object, long-running job), the description is remarkably complete. It explains what happens (validation, max_certainty compilation), what does not happen (fetching), the return values (search_id, compiled persona, next_step), and the next action (recall). The only minor omission is error handling, but that's not essential for selection and invocation. It fully equips an agent to know when and how to use it.

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 100% with detailed descriptions for both persona and label. The main description adds even more: it lists all scorable attributes (country, city, language, etc.), explains that some are advisory, and provides an example of the persona structure. This enriches the schema's semantics, helping the agent construct a valid persona and understand weight/required flags. The value added exceeds the baseline.

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 immediately states the verb and resource: 'Begin a persona search and get back a search_id to drive it with.' It also clarifies what it does NOT do ('Nothing is fetched here') and explains its role as the initialization step for a multi-stage search. This clearly distinguishes it from sibling tools like instagram_search_recall and instagram_search_users.

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 explicitly instructs the next step: 'After this, call instagram_search_recall with a probe plan.' It also explains the long-running nature and that it must be run stage by stage, implying this tool is the starting point. It differentiates itself from alternatives by stating it only validates the persona and opens a job, not fetching data.

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