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

Annotations already declare readOnlyHint=true and destructiveHint=false, but the description adds genuinely new behavior: it writes nothing to the CRM or internally, saves nothing, and previewing does not create or change a template. It also discloses the bounded-scan behaviour and the bound_reached signal, which is not derivable from any structured field.

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?

Front-loaded with the payoff (what the preview reports) before the mechanics, and every sentence carries information. Slight redundancy remains: the 'takes the scheme itself, not a template id' / 'saves nothing' points are restated in the schema and repeated with emphasis, costing a little efficiency.

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?

With no output schema, the description carries the full burden of describing return values and does so concretely: examined/changed/flagged counts, per-field change counts, before/after samples, and bound_reached. All four parameters are addressed, including the required connection_id path via list_crm_connections, leaving nothing an agent needs to call it correctly.

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 description coverage is 100%, so the baseline is 3. The description adds real meaning on top: it stresses that `config` is the scheme body rather than a template id (enabling unsaved/hypothetical schemes) and that `filter_spec` scopes the previewed population exactly as a scoped run would, which clarifies semantics the schema only gestures at.

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?

States a specific verb and resource ('preview a normalize scheme') and immediately enumerates what the preview returns: records examined, changed, flagged, per-field change counts, and before/after samples. It also distinguishes itself from the run-creating sibling by framing the output as 'What a normalize scheme WOULD do', making it separable from create_normalize_run without opening a schema.

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?

Gives explicit routing: pass the `config` from get_rule_template to preview a stored scheme, and call get_normalize_catalog for rules/options, and it explains filter_spec scopes 'exactly as a run would'. It stops short of explicitly contrasting against create_normalize_run ('use preview before committing a run'), so the when-not is implied rather than stated.

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