front_door

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

Annotations only declare readOnlyHint=true and openWorldHint=false. The description goes far beyond this: it lists assumptions (sequential exchangeability, positivity, correct specification), failure modes (stabilized weights with extreme values, and specific remedies like truncation, simplification, or TMLE), validation tiers (known-truth, reference, external-parity, Monte Carlo artifact), and a typical minimum N of 300. This rich behavioral context is not present in the annotations and adds real value for an agent deciding whether to trust results or handle failures.

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 dense but every sentence earns its place. It opens with the core purpose, then covers supported mediator types, validation, assumptions, pre-conditions, failure modes, alternatives, and sample size – all in a structured format with explicit labels (Assumptions:, Pre-conditions:, Failure modes:, Alternatives:). It is front-loaded with the most important information and avoids redundancy.

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 (12 parameters, output schema present), the description covers everything an agent needs to call it correctly: purpose, assumptions, pre-conditions, failure modes, alternatives, and typical sample size. The output schema handles return values, so the description need not repeat those. Nothing critical is missing for correct selection and 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 coverage is 100%, so the baseline is 3. The description adds interpretation beyond the schema: it explains that integrate_by controls Pearl (marginal) vs Fulcher et al. (conditional) aggregation, and it explains mediator_type supports binary or continuous mediator. These clarify the semantics of the enum parameters. It does not detail every parameter, but the schema already covers them, so the description's added meaning is valuable and pushes the score above baseline.

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 precise statement of what the tool does: 'Pearl's front-door adjustment: identifies ATE with unmeasured confounding when a mediator fully transmits the effect of D on Y.' It names the specific estimator (front-door adjustment), the causal estimand (ATE), and the condition (unmeasured confounding with a fully-transmitting mediator). It also distinguishes from sibling tools like frontdoor and mediation_decompose by naming the mechanism (Pearl vs Fulcher aggregation) and listing alternatives explicitly.

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 gives explicit when-to-use guidance: it is appropriate when there is a mediator that fully transmits the effect of D on Y in the presence of unmeasured confounding. It lists pre-conditions (sequentially measured covariates, treatment, outcome; models for treatment and outcome) and explicitly names alternatives: 'Alternatives: sp.tmle, sp.g_computation, sp.ipw.' This directly routes the agent to the correct tool and away from siblings.

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