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

Annotations confirm destructive/idempotent nature; the description adds valuable behavioral context not in annotations: the revert target state ('AI-inferred rubric') and operational cost ('Free'). No contradictions 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?

Three short sentences with zero waste: first states action and outcome, second specifies parameter provenance, third states cost. Efficiently front-loaded with the critical destructive action.

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 single-parameter destructive operation with no output schema, the description adequately covers the action, prerequisite data sources, and side effects (reversion to AI rubric). The annotations handle safety hints, so the description doesn't need to repeat them.

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?

With 100% schema description coverage, the baseline is 3. The description repeats the schema's explanation of model_id source (from create/list tools) without adding additional semantic context like format constraints or validation rules.

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 uses specific verb 'Clear' with specific resource 'rubric overrides for a custom eval model' and clarifies the end state ('reverting to the AI-inferred rubric'). This clearly distinguishes it from siblings like atlas_set_custom_eval_rubric_overrides and atlas_get_custom_eval_rubric.

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?

Provides clear context by specifying the model_id source tools (atlas_create_custom_eval_model or atlas_list_custom_eval_models) and notes cost ('Free'). Lacks explicit when-not-to-use guidance comparing it to set_overrides, but the revert behavior implies the use case.

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