prediction_calibration

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

With no annotations, the description carries full behavioral burden. It does add real value: input source (closed cards), grouping by kind+unit, the warning that small N is noisy and only trustworthy=true groups should be shown, and the interpretive note that forecast/fact divergence reflects model calibration rather than failure. It does not, however, disclose permissions, cost, rate limits, or whether any state is mutated (a read-only nature is only implied).

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?

Dense and front-loaded: the core action leads, followed by metric details, output filtering, and interpretation. Every clause carries information, though the sentence is long and semicolon-chained rather than cleanly broken.

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?

With no output schema, no parameters, and no annotations, the description is the whole specification and it covers what is computed, how results are grouped, how to present them, and how to interpret them. Minor gaps remain around access prerequisites and whether the operation is purely read-only.

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?

The tool takes zero parameters, so the schema has nothing to document and the baseline of 4 applies. The description adds no parameter meaning because none exists.

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 precise verb+resource (computes numerical calibration of forecasts from closed Decision Cards) and enumerates the metrics produced (Brier/log score for events, MAE/interval coverage for quantities). It also carves out its domain by contrasting with 'подачи', but never names the sibling tools it differs from (e.g. calibrate, calibrated_consult_journal), so sibling differentiation is implied rather than explicit.

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?

Implicitly scopes usage to 'closed Decision Card' inputs and gives a presentation rule (show only trustworthy=true groups), which tells the agent how to use the result. But it never states when to choose this over the many sibling calibration/consult/journal tools, nor any when-not condition or prerequisite.

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