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

Annotations already declare readOnly/idempotent/non-destructive/closed-world, so the safety profile is covered. The description adds real behavioral context beyond them: evidence is 'from registry-run benchmarks only', savings exist only for measured models, unknown fields are refused with the allowed list, and do_not_use/no_match are normal answers rather than errors. It stops short of stating latency, rate limits, or caching, but for a read-only resolver that is minor.

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?

Front-loaded with purpose and decision vocabulary, then when-to-use, exclusions, input and output in a fixed order; every clause carries information. It is dense and long for one paragraph, but little of it is waste.

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?

No output schema exists, yet the description enumerates the ak.resolve.v1 payload (decision, capability id, reason codes, measured saving with interval, baseline estimate with break-even, confidence, evidence label, price, free alternatives, next action, evaluation_id, detail tool). Combined with the named inputs and constraint semantics, an agent has everything needed to call and interpret it.

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 the enforcement behavior that the schema only implies ('any other field is refused with the allowed list') and groups the inputs as task + optional model/expected_runs/baseline_cost_estimate_usd/context/constraints. It doesn't add syntax beyond the schema, so it is a modest step above baseline.

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+resource: turn a plain-language task into a single decision word about whether a registry capability is worth using. It names the exact answer vocabulary (use, do_not_use, use_free_alternative, insufficient_evidence, blocked_by_constraints, no_match), so an agent can tell it apart from search_capabilities or evaluate_capability without opening a 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?

Explicit 'Use when' clause (you have a task and no capability name, and want one decision word) plus an explicit 'Not for' clause naming three alternatives and the condition that selects each: evaluate_capability for the full arithmetic, search_capabilities for a plain list, get_capability when the capability is already named.

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