sw_compare_arms

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

No annotations are provided, so the description carries full disclosure burden. It reveals important behaviors: pairing cancels common channel effects, fairness check blocks config drift and CSI inconsistencies, statistical tests use paired t-test and Wilcoxon signed-rank with specific blocking conditions (95% CI crossing zero, single sample contribution >50%), and it explicitly states when gates are not passed it will state the conclusion is invalid and why. This is rich behavioral context far beyond a typical tool description.

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 description is front-loaded with a one-sentence summary, then uses a bulleted list to explain the four unique capabilities, and ends with parameter-value notes. Despite being longer than most descriptions, each sentence/bullet adds distinct information (pairing benefits, fairness issues, test details, conclusion behavior). No fluff or redundancy.

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 the tool's high complexity (paired comparison, multiple gates, statistical tests) and the presence of an output schema, this description is nearly complete. It explains the method's purpose, the gate semantics, the conclusion behavior, and key parameter defaults. The only gaps are minor parameter meanings (receiver, max_samples) and interaction with other tools, but the description stands alone well.

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 input schema provides no descriptions (0% coverage) and no enums. The description compensates by explicitly documenting the allowed values for method_* (svd/svd_wideband/type1/dft/mrt/identity), the meaning of csi_* (ideal vs estimated), and the behavior of snr_db when omitted (uses per-sample SINR). This covers the core comparison parameters. However, it does not explain receiver, max_samples, dataset_id, or name_*, leaving some semantic gaps. Still, the value added for the central parameters is substantial.

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 clear statement: 'Run two schemes on the same batch of channels, do paired comparison, and pass gates 2 and 3.' It then enumerates four concrete capabilities (pairing, fairness check, statistical tests, conclusion sentence) that distinguish it from ordinary mean-comparison tools and likely from siblings like sw_compare_scenarios. This specific verb+resource+scope makes the purpose unambiguous.

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?

It states 'This is the last gate before drawing conclusions,' providing a clear temporal/decision context. It also implies alternatives by saying it does four things that ordinary 'mean comparison' cannot do, implying use when those advanced features are needed. However, it does not explicitly name sibling tools or give when-not-to-use exclusions, so it falls short of a perfect 5.

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