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

Adds essential behavioral context beyond annotations: the synchronous/long-running nature (30-60s), cost ('Free'), and mechanism ('analyzes all uploaded artifacts'). Does not contradict annotations (readOnlyHint: false aligns with active 'infer/generate' verbs). Minor gap: does not address idempotentHint: false implications (subsequent calls may generate different rubrics).

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?

Five sentences, zero waste. Front-loaded with purpose ('Infer...'), followed by mechanism, performance characteristics, parameter provenance, and cost. Every sentence delivers unique, non-redundant value not found in structured fields.

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 no output schema exists, the description adequately describes the conceptual output ('evaluation dimensions with scoring criteria') and operational context. Could improve by sketching the return structure or noting the non-idempotent behavior, but sufficient for a single-parameter inference tool.

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 guidance on model_id sourcing but adds no additional semantic details, examples, or format constraints beyond what the structured schema already provides.

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 specific verb-resource pair ('Infer an evaluation rubric') and scope ('from the model's artifacts'). It clearly distinguishes from siblings like 'atlas_get_custom_eval_rubric' (retrieve existing) and 'atlas_set_custom_eval_rubric_overrides' (manual update) by emphasizing the AI-driven generation aspect ('AI analyzes... generates').

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 critical operational constraints ('synchronous AI call that may take 30-60 seconds') and prerequisites ('model_id from atlas_create_custom_eval_model'). However, it lacks explicit guidance on when to prefer this over manual rubric creation via sibling tools or what prerequisites (uploaded artifacts) are needed for successful inference.

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