manski_bounds

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

The annotations only indicate readOnlyHint=true, so the description carries most of the behavioral burden. It does this well by disclosing that the output is an interval, that only weak assumptions are imposed, that bounds may be too wide, and that additional restrictions like monotone treatment response can tighten them. It does not contradict the read-only annotation.

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 compact and front-loads the core purpose before moving to assumptions, preconditions, failure modes, and alternatives. Every section has a purpose. The 'Validation: validated evidence tier...' line is somewhat boilerplate-like, but it does not materially bloat the description.

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 14 parameters, an output schema, and read-only annotations, the description covers the important contextual gaps: what identification strategy is used, that the result is an interval, when bounds are uninformative, what alternatives exist, and a minimum N guideline. The output schema handles return-value details, so no major missing element blocks correct invocation.

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 parameters are already well documented in the schema. The description adds high-level context around assumptions and constraints rather than restating parameter meanings, which is appropriate for this coverage level. It does add a useful hint about binary selection/attrition indicators, but this is mostly relevant to Lee bounds rather than this tool.

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 verb and resource: 'Compute Manski (1990) worst-case bounds on ATE.' It further clarifies that the result is an interval rather than a point, which sharply distinguishes it from point-identifying estimators. It also names the related Lee and Oster bounds, so an agent can tell them apart without inspecting schemas.

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 meaningful context: only weak set-identifying assumptions are imposed, bounds may be too wide, and alternatives such as Lee bounds and Oster's delta are called out with their distinguishing assumptions. It does not state an explicit 'use this when... / use that when...' rule, but the preconditions and failure-mode guidance make selection reasonably clear.

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