calibrate

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

With no annotations, the description does real work: it discloses the side effect that calibration.json is written into the run folder, and names the verdict categories (trust / correct / walk-away) and the AIPW correction. It omits failure behavior (missing run, malformed CSV) and permission/overwrite semantics.

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?

One dense, front-loaded sentence that leads with the core verb and ends with the write side effect. The long parenthetical on CSV format is justified, though the sentence is heavy to parse.

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?

No annotations and no output schema, so the description carries full load; it covers the deliverable file and headline metrics but leaves three parameters and error conditions undefined. Adequate to start, not complete for a 5-parameter tool.

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 0% and three of five parameters (by, map, field) are never mentioned, so an agent cannot know what they control or their formats. Only real_csv is explained in useful detail (columns, label encoding, row count) and 'run' is implied by 'run folder'.

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 a specific verb and resource: 'Check a run's simulated survey against real answers', and enumerates the concrete outputs (covariate overlap, per-audience calibration, verdict, AIPW-corrected rate). This clearly separates it from siblings like results or actuals_import.

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?

Implies the tool is used after a run exists and that you supply an external ground-truth CSV, and even gives sizing guidance ('50-300 rows is enough'). But it never says when to prefer this over siblings such as results, backtest, or actuals_import, nor any prerequisites.

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