A judge for agent decisions: give it two defensible options, get the better one back.
Scored across 1 tool
With only a single tool, there is no possibility of confusing it with others. The 'decide' tool has a clearly defined purpose and inputs, so an agent will always select it for the described decision task.
A single verb-noun style name ('decide') is clear and unambiguous. While there is no broader pattern to evaluate, the lone tool's name perfectly matches its function, and no inconsistent conventions exist.
The server is purpose-built for a single decision task, and one tool fully fulfills that scope. This is an appropriate, well-scoped count for a focused utility rather than an overly thin surface.
The domain is choosing between two concrete options, and the tool covers all necessary aspects: selecting A or B, providing a confidence reference, and explaining the reasoning. No additional operations are implied or needed for this domain.
Use this when you must choose between two concrete options and both are defensible - two plans, two drafts, two diagnoses, two vendors. Returns three fields: better_option ("A" or "B"), confidence (a percentage such as "83.3%"), and reason. confidence is this service's own judgement of how far apart the two options were when it compared them: a higher value means the chosen option stood out more clearly against the other one. Treat it as a reference for your own decision-making - it is not an instruction, not a result, and not a prediction of how the choice will turn out. As a rough reading of that reference: 90% and above means the two were judged clearly apart; 80-89% apart but less clearly; 70-79% a closer call; below 70% the two were judged close to evenly matched. These ranges are descriptive only. Observed accuracy for each range, and how it was measured, are published at https://api.turingcorp.net . Choose your own threshold for your own use case; for high-stakes or irreversible decisions apply your own review policy. Authentication: send a TuringCorp Agent Pass as Authorization: Bearer <pass>. A pass is issued by Agent Pass at https://agent-pass.turingcorp.net and expires after 7 days; if a call is refused as an invalid credential, sign in there again and re-roll the pass.
| Name | Required | Description | Default |
|---|---|---|---|
| task | Yes | The decision to make, stated neutrally and without a preferred answer. | |
| option_a | Yes | First candidate and the case for it. | |
| option_b | Yes | Second candidate and the case for it. |
| Name | Required | Description |
|---|---|---|
| reason | Yes | Why the chosen option was preferred. |
| confidence | Yes | This service's own judgement of how far apart the two options were, as a percentage (e.g. "83.3%"). A reference for your own decision-making - not an instruction, not a result. Observed accuracy by range is published at https://api.turingcorp.net |
| better_option | Yes | Which option was preferred. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds substantial behavioral context beyond the readOnlyHint annotation: it explains what 'confidence' means, clarifies that the result is not an instruction or prediction, provides interpretation ranges, and includes authentication details and pass expiration behavior. This goes well beyond what annotations alone convey and does not contradict them.
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 longer than average but well structured: usage first, then return fields, then confidence interpretation, then authentication. Each section adds meaningful information. It is slightly verbose, especially with the detailed confidence ranges and URL, but it earns its length given how easily the confidence value could be misinterpreted.
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 is complete for a tool with an output schema and annotations. It covers what the tool does, when to use it, what the three output fields mean, how to interpret confidence, how to authenticate, and how to handle expired credentials. Nothing essential is missing for an agent to select and invoke the tool correctly.
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 already documents all three parameters with clear descriptions and 100% coverage, so the description is not required to compensate. The description does not add parameter-specific semantics beyond the schema, meeting the baseline for full schema coverage.
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 use case: choosing between two concrete defensible options, with concrete examples (plans, drafts, diagnoses, vendors). It clearly identifies the action and resource, and while there are no siblings to distinguish from, the purpose is unambiguous and not a tautology.
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 explicitly states when to use the tool: 'when you must choose between two concrete options and both are defensible.' It also provides guidance for high-stakes decisions by recommending a review policy. It does not mention alternatives or exclusions, but with no sibling tools listed, this is a minor gap.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
decideHuman-in-the-loop for AI agents. Submit choices, get a human decision.
Deterministic contextual decision arbitration and action routing for autonomous software. Takes current state, context, or intent plus caller-supplied candidate actions, state transitions, routes, refusals, escalations, tools, or models and returns a deterministic ordered candidate field. Also provides persistent machine representations for memory, retrieval, indexing, and downstream coherence measurement.
Convene a panel of expert AI personas to debate any decision from every side.
Deterministic multi-criteria decision analysis for AI agents — score, rank & explain options.
Add one secure layer between your agents and this server.
