offline_safe_policy

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

Annotations declare readOnlyHint=true, and the description adds valuable behavioral context beyond that: the method is 'safe' (pessimistic), it enforces a cost constraint, and it warns about poor coverage with a suggested mitigation. It does not explicitly state that it returns a fitted result, but the output schema and as_handle parameter cover that. The description adds substantial behavioral insight without contradicting the read-only hint.

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 compact yet information-dense: it leads with the core purpose, then follows a logical flow of assumptions, preconditions, failure modes, alternatives, and a sample-size heuristic. Every sentence contributes value; there is no filler. The structure makes it easy for an agent to scan quickly and extract the essential conditions.

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 method with nuanced assumptions, the description covers them thoroughly: statistical assumptions, data requirements, failure modes with mitigation, and typical N. Combined with the full schema (14 params, all documented) and an output schema, nothing critical is missing. The description is complete enough for an agent to decide when and how to invoke it correctly.

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 every parameter is documented in the schema. The tool description itself adds no parameter-specific meaning—it focuses on method assumptions and usage. This meets the baseline of 3; the description does not need to compensate for missing schema documentation, but it also does not enrich parameter understanding beyond what the schema already provides.

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 a specific verb-resource pair: 'Safe offline policy learning with a cost-constraint.' It clearly distinguishes this tool from many siblings by naming explicit alternatives (sp.offline_safe_policy, sp.policy_value) and framing the method as safety-focused. The purpose is unambiguous and sets it apart from the large sibling set.

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 explicitly lists assumptions (sequential ignorability, positivity, Markov dynamics), pre-conditions (logged trajectories from a known/estimable behavior policy), failure modes (poor coverage) with a remedy, and alternatives. This is exemplary guidance: it tells the agent exactly when to use this tool, what must hold, and what to do if assumptions fail.

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