hal_tmle

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

Annotations provide readOnlyHint=true and openWorldHint=false, but the description goes well beyond them by detailing known limitations, failure modes, the no-op status of the projection variant at v1.11.x, and the parity-test gates. The only minor gap is the truncated failure-mode sentence, which cuts off before completion.

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 and structured with labeled sections (Validation, Known limitations, Assumptions, Pre-conditions, Failure modes), making it scannable. It earns its length because it conveys serious limitations and assumptions. It loses one point because the final sentence is truncated mid-thought, and some redundancy exists around the projection variant.

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 — 17 parameters, two variants, one unusable variant, exposure to assumptions and failure modes — the description covers the essential context well and even points to a design doc. It stops short of a full account of expected outputs, but the output schema exists and the description explicitly names failure modes. The truncation of the failure-modes sentence prevents a 5.

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 schema already documents all 17 parameters. The description adds very little parameter-level meaning beyond what the schema provides, though it does contextualize the 'projection' variant's relationship to the implementation roadmap. Baseline 3 is appropriate.

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 'TMLE with Highly Adaptive Lasso (HAL) nuisance learners', naming a specific statistical method, and names two variants ('delta' and 'projection') with their behaviors. It also cites Qian & van der Laan 2025, distinguishing this from siblings like tmle, ltmle, and super_learner.

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 does not explicitly name sibling tools, but it gives strong contextual guidance: it warns that variant='projection' raises NotImplementedError and directs users to docs/rfc/hal_tmle_projection.md, and it lays out assumptions/pre-conditions. It lacks explicit 'use X instead of Y' statements, but for a causal-inference estimator with an obvious family, the implied usage is clear.

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