rca_analysis_ensemble

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

The description discloses the algorithm steps (dispatch_rca, weighted scoring, normalization), the Pro+ feature requirement, the default save=true persistence behavior, and possible error types (bad_input/not_found). Annotations are all false and thus provide little behavioral signal, so the description carries the burden and handles it well. It could additionally caution about the side effect of persisting results by default, but the save parameter is at least mentioned.

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 organized with a front-loaded purpose, usage guidance, a numbered algorithm, structured Args, and a Returns section. There is minor redundancy between the Args section and the schema descriptions, but the length is justified by the algorithm complexity. No filler words or vague statements.

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?

The description covers when to use, the algorithm, the 2-5 model constraint, the shared payload requirement, default save behavior, output JSON structure, and error cases. An output schema exists, so detailed return-value documentation is not required. The only slight gap is not elaborating on payload validation across model families, but it acknowledges this with 'shape depends on the models' families.'

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 Args section enriches the schema meaningfully by noting weights must be 'same length as model_ids,' defaulting to equal weighting, and describing payload as 'shared input, shape depends on the models' families.' The schema already provides solid descriptions for all fields, so this is additive rather than necessary, but it adds useful constraint context. Token and client_id are left to the schema, which is sufficient.

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 states a precise operation: 'Run 2-5 different RCA models on the SAME payload and combine their root-cause scores via weighted voting.' It is a specific verb (run/combine) with a clear resource (RCA models/payload), and it explicitly contrasts with rca_analysis_batch by saying 'for the SAME model run over multiple different incidents instead, use rca_analysis_batch.' This makes the tool's purpose unmistakable and distinguishes it from siblings.

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?

The description gives explicit guidance: 'Use this when you're unsure which single model family fits the data best and want cross-validation across families,' and it names the exact alternative tool and condition. This is a clear when-to-use and when-not-to-use statement, leaving nothing to inference.

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