company_calibrate

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

With no annotations, the description carries the full burden. It explicitly states 'Proposals only: nothing is applied until the owner writes them into the bundle' and 'Deterministic; read-only.' It also discloses the minimum observation threshold (default 6) and that proposals are declined below it. This is thorough for a read-only operation, though it does not mention auth requirements or rate limits.

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 well-structured and front-loaded with the core function. Each sentence adds substantive detail: the percentile rules, the observation minimum, the proposal-only nature, and determinism. It is reasonably concise given the richness of information, though it could be tightened by removing the repetitive 'proposals only' phrasing.

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?

The description covers the behavioral context thoroughly (read-only, deterministic, proposal-only, minimum observations) and the output schema exists to describe return values. However, it is incomplete regarding how to invoke the tool: it does not explain the purpose of project_id, bundle_json, payload_json, or the operation parameter. An agent would need to guess the expected input structure, especially since schema coverage is nil.

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%, so the description must compensate. It only mentions 'payload minimum_observations, default 6,' which hints at a parameter within payload_json, but it does not explain the required project_id, bundle_json, or other fields. An agent would have to infer what to pass beyond the single hint, leaving the parameter semantics largely unexplained.

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 clearly states the tool's purpose: to plan settings based on the company's own history using plain percentiles. It gives specific examples of settings (customer concentration ceiling, minimum cost share, receivable-days target) and their calculation rules, making the function unambiguous. However, it does not explicitly differentiate from sibling tools like company_what_if or company_simulate, though its unique role is apparent.

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 guidance on when to use this tool versus alternatives. It does not mention conditions like 'use this when you need to propose calibration settings' or contrast with similar tools. The only usage-related note is that proposals are not applied until the owner writes them, which is behavioral rather than selection guidance.

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