feols

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

Annotations only declare readOnlyHint=trueusters. The description adds substantial behavioral context: computational complexity (dense O(n^2) for conley), concrete memory estimates (~0.8 GB at n=10,000, ~157 GB at n=140,000), statistical assumptions (strict exogeneity, no perfect collinearity), singleton handling, and failure modes. This goes well beyond the annotations and helps the agent anticipate the tool's runtime and correctness requirements.

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 long but well-structured with clear sections (Validation, Do NOT use when, Cost, Assumptions, Pre-conditions, Failure modes). It front-loads the core purpose, then organizes warnings and technical context logically. Each section earns its place; while condensed, it remains actionable for a complex 20-parameter tool.

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 high-dimensional fixed effects estimator with an output schema, the description is remarkably complete: it covers validation evidence, cost and memory behavior, assumptions, pre-conditions, failure modes, and guidance on alternatives for the conley case. The presence of an output schema means return-value details are already structured, and the description fills the behavioral and contextual gaps.

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 enriches the vcov parameter with cost/alternative guidance and adds context to the fml parameter about IV requirements. However, it references 'drop_singletons' while the schema defines 'fixef_rm', creating a naming mismatch that could confuse an agent. This prevents it from scoring higher.

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 'Estimate OLS / IV with high-dimensional fixed effects via pyfixest', which is a specific verb, resource, and method. It clearly distinguishes this tool from generic regression siblings like 'regress' and from Poisson/FE variants such as 'fepois'. It also includes validation parity evidence, reinforcing the purpose.

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 'Do NOT use when' clause explicitly warns against using vce='conley' on large datasets and names the preferred alternative (sp.conley). Pre-conditions further define appropriate contexts (long-format data, categorical FE columns, instrument sufficiency). However, it does not explicitly name sibling tools for the main use case, leaving some routing implicit in the purpose statement.

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