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

Annotations indicate `readOnlyHint=false` (a write operation) and `destructiveHint=false`; the description's 'Create' action aligns with this. It adds behavioral nuance by describing the `estimate_only` mode, which alters the operation to a non-creating preview and returns `expectedNumberOfContacts`. It also specifies the constraint that `name` is required and must be under 50 characters. This goes beyond the annotations and gives useful execution detail, though it does not mention potential idempotency or duplicate errors.

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 long (about 20 lines) but organized with headers ('WHEN TO USE', 'PARAMETERS', 'RETURNS', 'IMPORTANT NOTES'). It is front-loaded with the purpose and key mode. However, it includes redundant sections like 'ALSO KNOWN AS' and 'KEYWORDS' that add noise, and the parameter section largely duplicates schema descriptions. The overall structure helps navigation, but the length and redundancy prevent a higher score.

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 eight parameters (one required) and no output schema, the description covers the essential aspects: purpose, usage scenarios, parameter details, the `estimate_only` mode, and the name constraint. It also states the return type ('audience details with ID, status, and creation info'). However, it does not explain how multiple criteria are combined (e.g., AND vs. OR), any limitations on array sizes, or error conditions. This leaves gaps that could impact correct invocation for complex audience builds.

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 already carries a 100% description coverage, including enums for `employees` and `revenues`, a default for `estimate_only`, and a comprehensive list for `location_country_ids`. The description repeats this information with examples for each parameter, but effectively adds little meaning beyond restating schema details. For instance, the valid employee ranges are already in the schema, so the description's repetition is redundant. Baseline 3 is appropriate since the schema does the heavy lifting.

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+resource: 'Create a LinkedIn native criteria audience using LinkedIn-supported targeting criteria.' It explicitly lists the types of criteria (job titles, skills, company names, employee count, revenue, geography) and distinguishes this tool from other audience-creation siblings like `create_facebook_native_criteria_audience` and `create_linkedin_engagement_retargeting_audience`. The purpose is unambiguous and differentiated.

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?

A dedicated 'WHEN TO USE' section enumerates concrete scenarios: targeting by job titles, skills, company names, employee count, revenue, and location. It also explains the `estimate_only=true` mode as a preview option, which clarifies a distinct use case. However, it does not explicitly state when NOT to use this tool versus alternative audience-creation tools (e.g., `create_audience_from_segment`, `create_firmographic_audience`), so the guidance is strong on positive usage but lacks exclusion criteria.

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