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

With no annotations, the description carries real weight: it discloses idempotency semantics ('exact retries use the same idempotency_key') and a critical scope boundary ('No refund, payment cancellation, guarantee or money movement is authorized'), preventing an agent from misreading the refund_review remedy as an actual payout. It still omits role enforcement, valid state transitions, and error behavior, keeping it short of full disclosure.

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?

Four compact clauses, front-loaded with the workflow and followed by idempotency and safety constraints; every sentence earns its place. The opening sentence is slightly cryptic as an opener because it never names the tool's operation.

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 multi-action mutation tool with a oneOf schema, no annotations and no output schema, the description should at least map which actions require which field set and confirm that escalation/resolution are terminal state changes. It covers the lifecycle shape and safety boundaries but leaves the action-to-payload mapping entirely to the schema.

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 polymorphic fields are self-documenting; the description adds value by telling the agent what to do with idempotency_key on retries, which the schema only types as a UUID. It does not explain the action-to-branch split (open needs submission_id/criterion/evidence/remedy; respond/escalate/resolve need issue_id/message).

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 narrates the lifecycle (buyer opens, supplier responds, buyer escalates or resolves) rather than stating a verb+resource like 'Create or advance a delivery issue.' An agent can infer the tool mutates delivery issue state, but the operation itself is never named explicitly, and no sibling (e.g., get_delivery_issues, check_delivery_refund) is distinguished.

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 actor mapping (buyer opens, supplier responds, buyer escalates/resolves) gives implied guidance on who invokes which action, which is useful. But there is no explicit when-not guidance, no reference to alternatives like get_delivery_issues to find an issue_id or check_delivery_refund to verify refund eligibility, and no preconditions (issue state, ownership) for escalation.

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