partial_identification

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

Annotations declare readOnlyHint=true and openWorldHint=false, so safety is covered. The description adds behavioral context: the result is an interval not a point, and failure modes are described (bounds too wide). This goes beyond annotations and is valuable for setting expectations. 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 a single structured paragraph with labeled sections (Assumptions, Pre-conditions, Failure modes, Alternatives, Typical minimum N). It is dense but each sentence provides distinct information. Purpose is front-loaded. Could be slightly more concise, but overall well-organized and informative.

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 complex tool with 13 parameters, it covers assumptions, pre-conditions, failure modes, alternatives, and a typical minimum N. An output schema exists, so return format is likely covered. It does not explicitly link method values to the named alternatives, but the description mentions Lee bounds and Oster's delta. Overall, it is fairly complete for an agent to plan calls.

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 each parameter has a description, but many are terse (e.g., 'd parameter (str).'). The description adds pre-conditions (weakest credible identifying restriction, binary selection indicator for Lee bounds) which clarify selection and assumptions parameters. It does not elaborate on method or detail, but the schema already has some info. Baseline of 3 is appropriate; description adds a bit but not comprehensive.

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?

States 'Partial identification of ATE' clearly, a specific verb+resource. It describes the result as an interval and mentions assumptions (Lee bounds, Oster's delta) and alternatives, which helps distinguish from sibling tools like oster_delta, lee_bounds, manski_bounds. However, it does not explicitly state how it differs from those siblings beyond listing them as alternatives, so it is clear but not maximally differentiated.

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?

Provides explicit alternatives (sp.oster_delta, sp.lee_bounds, sp.manski_bounds) and a failure mode remedy (add a credible auxiliary restriction to tighten bounds). Pre-conditions and typical N are given. It lacks an explicit 'use this when you only have weak assumptions' statement, but the context strongly implies it. Overall, solid usage guidance.

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