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

Annotations already declare a safe read-only, idempotent, closed-world operation, and the description adds material context beyond them: 'these tools never charge', 'Every purchase still needs human approval', capabilities outside constraints are 'reported as blocked, with the reason code', and 'do_not_recommend is a normal answer' (guarding against an agent retrying on a negative verdict).

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?

Well front-loaded: purpose, then when, then not-for, then input, then output. It is long and dense (the output enumeration and 'the four evidence kinds listed apart' are heavy), but nearly every clause carries routing or contract information, so the size is largely earned.

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 an 8-parameter tool with nested objects and no output schema, the description supplies the missing return contract (ak.evaluation.v1 fields, ak.offer.v1, blocked statuses and reason codes) and the money-estimate input combination. An agent has everything needed to call it and interpret the result.

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 100%, so the baseline is 3, but the description adds real meaning: model/expected_runs/baseline_cost_estimate_usd are jointly required 'for a money estimate', and omitting capability makes the registry weigh its closest matches rather than failing. That interaction logic is not visible in the schema. It does not, however, explain the context-keying or budget_policy precedence beyond the schema's own text.

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 (decide) and object (whether a registry capability is worth using or buying), and pins the scope to registry-run benchmarks only. The description explicitly differentiates itself from four named siblings, so an agent can route without opening any schema.

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?

Contains an explicit 'Use when' clause plus a 'Not for' clause that names the alternative tool for each adjacent case: resolve_task for a bare answer, search_capabilities for a list, compare_capabilities for 2-5 candidates, estimate_roi for break-even without a task. It also clarifies the optional-capability branch (name one vs. let the registry weigh closest matches).

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