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

No annotations exist, so the description carries the full burden, and it delivers real behavioral context: read-only ('只读'), no authentication required ('免鉴权'), plus the data-integrity rules (sim and real strictly separated, only control-group numbers admitted, unreported metrics stored as null rather than extrapolated, sample sizes absent). It does not describe the response shape or pagination, which is the remaining gap.

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 core purpose is front-loaded in the first clause and the three disciplines are numbered, making the paragraph scannable despite its length. Some of the example questions border on redundant with the stated purpose, but every sentence carries operational content.

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 and no annotations, the description is the agent's only source, and it covers source provenance, exclusions, null handling, and access requirements well. It stops short of describing the record fields returned (model, metric, sim/real flag, source citation) and any result limits, which would be the last thing needed to call it with full confidence.

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 tool takes zero parameters, so per the rubric the baseline is 4; there is no parameter surface for the description to clarify or compensate for.

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+resource ('查询行为证据') and pins down the exact payload: open-source VLA real-robot benchmark success rates sourced from papers/official reports rather than self-generated logs. The scope is unambiguous and the example questions ('GR00T 真机成功率多少', 'SmolVLA 有实测数据吗') make the retrieval target concrete, but no sibling tool (e.g. explain_evidence_cohort, explain_evidence_mdv) is named to differentiate it from adjacent evidence tools.

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?

Explicitly states when to use it by enumerating the three example questions it answers, and states when NOT to rely on it: '样本量官方均未报告 ⇒ 不可用于统计比较' rules out statistical comparison, and the sim-vs-real discipline plus the control-group rule define which data qualifies. This is when/when-not guidance of an unusually high standard.

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