Research ads

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

Annotations provide readOnlyHint, openWorldHint, and destructiveHint=false, but the description goes beyond by disclosing that this is an agentic loop taking 30-60 seconds, that it spends credits (a handful vs. a single-call cost), and that it performs several rounds of library pulls plus a written synthesis. This behavioral context, including cost and latency, is critical and not available in the annotations, so the description fully carries its burden.

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 front-loaded with the core purpose and examples, then provides usage guidance and caveats in a logical flow. Every sentence earns its place—no filler, clear structure, and the critical warning ('WRONG tool for a question that names its own answer') is highlighted. It is appropriately length for the tool's complexity.

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 tool that is an agentic loop with no output schema, the description explains the behavior (multi-round, synthesis), the time cost, and the credit cost, and routes to faster alternatives for simpler queries. It doesn't explicitly describe the output format beyond 'written synthesis', but that is sufficient for an agent to know what to expect. The context is adequate for deciding when to invoke this tool, and the only minor gap is the absence of a precise return structure, which is acceptable given the open-ended nature.

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 input schema covers both parameters (query and brand) with clear descriptions, so the baseline is 3 per the rubric. The description adds an example of a query ('the longest-running protein-pancake ads on Meta') and implies that the query should be open-ended, but it doesn't explain the brand parameter beyond what the schema provides. Since schema coverage is 100%, the description's added semantic value is marginal, though it does reinforce what kinds of queries are appropriate.

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 'Open-ended ad research that needs JUDGMENT across platforms' and gives concrete examples ('comparisons', 'what angle is working', 'who else is doing this'), making the tool's specific purpose unmistakable. It also names two sibling tools (pull_competitor_ads and search_meta_ads) and explicitly contrasts them, so an agent can distinguish this from the rest of the tool list.

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 states when to use this tool (open-ended questions where the right sources are unknown) and when NOT to use it ('a question that names its own answer'). It names the precise alternatives—pull_competitor_ads for a named brand's live ads, search_meta_ads for one keyword or advertiser on Meta—and notes they are single calls that return quickly, leaving no ambiguity about routing.

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