help_assist

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

With no annotations, the description carries the full burden of behavioral disclosure. It openly states that false is illegal, omission is cannot-adjudicate, and that fictional possibility remains an authorized DM ruling (including agent-DM usage surfaced as data.gm_discretion). It also discloses the return shape ({grants_advantage}). These are non-obvious behaviors beyond the schema.

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 four sentences, front-loaded with the core purpose, then logically structured by kind. Every sentence adds essential information without padding. It packs complex rules into a dense but readable format.

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?

Despite many infra parameters (table_policy, asserted_facts, policy_context), the description focuses on the domain logic that an agent must know to invoke correctly: required facts and their constraints. It also notes the DM discretion mechanism. It does not explain the standard policy/context fields, but these are likely common across sibling tools and the output schema is present, so the description is largely complete for selection and invocation.

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 only 25% (two booleans have descriptions), but the description adds significant semantic context: it explains that the boolean parameters must be explicit facts, false is illegal, and omission leads to cannot-adjudicate. It also clarifies the 'kind' enum and the meaning of the return value. This goes beyond the schema's bare types and enum.

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: 'Adjudicate the Help action' with an SRD citation. It then breaks down the two supported kinds (ability-check and attack-roll), clearly distinguishing this tool from siblings like check_make or attack_modifiers. The scope is unambiguous.

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 tells when to use the tool: when adjudicating a Help action, and for each kind it states the explicit fact prerequisites (helper_has_relevant_proficiency for ability checks, enemy_within_5ft for attack rolls). It also explains illegal and cannot-adjudicate cases, implying the tool is only usable under those exact conditions. It does not name alternative tools, but the conditions are specific enough.

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