arbitrate_consensus

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

With no annotations provided, the description carries the full burden of behavioral disclosure. It does mention deterministic behavior, a sub-5ms latency, and a cost figure, which are useful. However, it fails to disclose what the output looks like (despite an output schema), how it handles edge cases (e.g., ties, insufficient candidates), whether the operation is read-only or has side effects, and any failure modes. The description reads more like a promotional pitch ('Eliminates expensive LLM-as-a-judge spans') than a transparent behavioral contract.

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 two sentences with zero filler. It front-loads the cost and core purpose, then adds a performance/benefit statement. Every word earns its place, and it is appropriately sized for the tool's complexity.

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 tool with six parameters, an output schema, and a non-trivial consensus algorithm, this description is incomplete. It does not explain the required 'candidates' format, the meaning of the optional keys, how the agreement_threshold affects results, or what the output contains (though an output schema exists, the description does not bridge the gap between inputs and outputs). An agent would struggle to correctly invoke this tool without probing the schema fields.

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?

Schema description coverage is 0% – the input schema provides only titles, no descriptions. The tool description does not explain any of the six parameters, though it vaguely hints at 'decision alignment' (decision_key) and 'numeric convergence' (numeric_keys). It does not clarify the purpose of entity_keys, payment_signature, agreement_threshold, or the structure of the candidates array. This is insufficient compensation for the low coverage, leaving the agent to guess parameter semantics.

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 clearly states a specific action: 'deterministically arbitrate consensus across multiple model outputs', with a clear resource and context. It also hints at the mechanism (detecting decision alignment, numeric convergence, dissenting outliers) which differentiates it from generic aggregation. However, it does not explicitly contrast with a sibling tool like 'random_decision_oracle', though it does mention eliminating LLM-as-a-judge, which is a conceptual alternative. This is strong but not a full 5 because it lacks a direct named sibling comparison.

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 implies usage when you have multiple model outputs and need a fast, deterministic consensus, and suggests it replaces expensive LLM judgment. However, it does not provide explicit guidance on when NOT to use it (e.g., when inputs are subjective or when nuanced judgment is required) nor name specific alternative tools. The context is clear but the exclusions are left to inference.

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