scpi

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

Annotations already mark the tool read-only, and the description adds meaningful behavioral context: it discloses that the tool runs simulations on top of a point fit, that runtime is dominated by simulation count, and that cores parallelizes the work. It also surfaces assumptions (SUTVA, no anticipation) and failure modes (large RMSPE, placebo non-extremity), which go well beyond the structured annotations. No contradiction with annotations.

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 well-organized with labeled sections (Cost, Assumptions, Pre-conditions, Failure modes, Alternatives) and front-loads the purpose. However, it is long, and the final sentence is truncated ('Typical minimu...'), which is a structural defect. Most content earns its place, but the truncation and slight redundancy around failure modes prevent a higher score.

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 complexity (19 parameters, 6 required, output schema present), the description covers cost, assumptions, preconditions, failure modes, and alternatives, so an agent has enough to select and invoke it correctly. The schema fills parameter details and the output schema handles return values. The truncated final sentence and the lack of an explicit rule for choosing this over point-estimate synthetic control siblings keep it from being fully complete.

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 100%, so the baseline is 3. The description only adds marginal context for cores ('parallelises it') and the general simulation step, but it does not explain the key interval-related parameters such as pi_type, e_method, or w_constr beyond what the schema already provides.

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 identifies the output type ('Prediction intervals') and the resource ('synthetic control methods'), so an agent can infer the tool's purpose. It is phrased as a noun phrase rather than an explicit verb+object command, but the name and content make the action clear. It does not explicitly contrast with sibling synthetic-control point-estimators, though 'prediction intervals' differentiates it.

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 provides explicit pre-conditions, assumptions, failure modes, and an alternatives list, giving an agent strong guidance on when results are valid and what to do if assumptions fail. It does not state a crisp decision rule such as 'use this when you need interval estimates rather than point estimates', so some inference is still required.

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