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

Adds real behavior beyond the annotations: live re-evaluation on every use ('a batch or list sent to it reaches whoever matches then'), the 10-rule ceiling, and the 20,000-event search window. It omits the persistence/irreversibility angle that destructiveHint=true implies, and does not mention that name is optional, but the dynamic-membership disclosure is exactly the non-obvious fact an agent needs.

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?

Three sentences, purpose front-loaded, and the long middle sentence is a dense but scannable enumeration with clear 'Attributes:' / 'Behaviour:' labels. Some fillers ('a segment, a cohort') and the run-on rule listing could be tightened without losing information.

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 with no output schema, the description never says what comes back (segment id? object?) or whether the name must be unique, and it does not touch the optional name/description params despite required-parameters being 0. Rule construction, the highest-complexity part, is covered adequately, so the gaps are moderate rather than disqualifying.

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 description explains the rule vocabulary well: 'field = "field", key = the name', batch key = batch id with its five values, time values as ISO dates or windows like 7d, and 'all (or any) of up to 10 rules' maps to match and rules. However, the top-level name and description parameters are never addressed, and the per-field semantics largely duplicate what the nested schema descriptions already say.

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?

Opens with a concrete verb and resource ('Save an audience (a segment, a cohort)') and states where it surfaces (Audience → Segments) and what consumes it (broadcasts). It never names the nearest siblings (create_list, query_audience, preview_segment), so differentiation is implied by the 'one rule' vs 'rules' framing rather than stated.

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?

Usage is only implied: 'broadcasts can be sent to it' and 'evaluated live each time it is used' suggest when a saved segment is preferable to a one-off query. There is no explicit when-to-use / when-not guidance, and the alternatives an agent would hesitate between (create_list, query_audience, preview_segment) are never mentioned.

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