irf

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

Annotations declare readOnlyHint=true, and the description's 'Compute' aligns with a read/compute operation, so there is no contradiction. The validation sentence attempts to mention 'validated evidence tier' but is vague and does not explain what the tool actually does with that validation, how it reports it, or any behavioral nuances such as defaults for impulse/response/orthogonalization. It adds little beyond the schema's parameter descriptions.

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 first sentence is concise and front-loaded, stating the core purpose. However, the second sentence about 'Validation: validated evidence tier' is jargon-heavy and arguably waste, since it doesn't specify what evidence tier applies, how validation is performed, or what the agent should do with it. A two-sentence description could be tight, but the second sentence earns little of its place.

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 tool with 11 parameters, an output schema, and full parameter descriptions, the description is passable but leaves gaps. It does not explain that var_result refers to a result from a previous var call, nor does it describe any chaining or ordering context (e.g., fit VAR first, then call irf). The validation sentence hints at some evidence tier but is too vague to be actionable. The output schema exists, so return-value documentation is not required, but usage context is thin.

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 all parameters are individually documented in the input schema. The description itself adds no parameter-level meaning beyond the schema, which is acceptable under the baseline rule. It does not clarify the relationship between var_result, result_id, and data_path, or how as_handle chaining works, but the schema covers each parameter individually.

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 begins with a clear verb+resource: 'Compute impulse response functions from VAR.' This effectively identifies the operation on a specific model type, distinguishing it from the many sibling tools. However, it lacks explicit differentiation from related time-series tools such as local_projections or var, but the phrase 'from VAR' is reasonably specific.

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 no explicit guidance on when to use this tool versus alternatives like var, local_projections, or granger_causality. It implies usage through the name and first sentence, but it does not state prerequisites—such as first fitting a VAR with the var tool—or explain when one might instead use local_projections for impulse responses. The validation sentence is unclear and doesn't help with tool selection.

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