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

Annotations already declare destructiveHint=true and readOnlyHint=false, and the description adds substantial context beyond them: it cannot act on behalf of another person, it does not sign documents, it names canonicality (trust returned actual_tool_output instead of a state-recovery call), and it gives idempotency/retry guidance. For a destructive write this is unusually complete 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?

The key behavioral rules are front-loaded, but the trailing Canonicality/Idempotency/Confirmation boundary block reads as generic boilerplate that could apply to many tools and dilutes the tool-specific content. It is information-dense rather than padded, but the template-like tail costs structure points.

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 is strong on process, consent, and safety, which suits a destructive mutation with no output schema. What it omits is the shape of the nested identity payload it is expected to save: which fields are required, how address should be collected, and how the tax/immigration fields relate to the disclosure. For a tool whose only output is a saved identity object, that gap is notable.

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?

With schema description coverage reported at 0% for the top level, the description must carry the burden, and it only partially does. It clarifies that identity.name is the full legal name as on government ID and that consent means explicit plain-text agreement rather than a choice option, which is genuinely useful. However, joinCode, address, usTaxpayer, usCitizenOrPermanentResident, immigrationStatus, and f1WorkAuthorization receive no explanation, leaving a nested required object largely undocumented in prose.

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 states a specific action on a specific resource: show the invitee the disclosure from review_invited_identity and then save their own approved identity details. It positions itself relative to redeem_invite (prerequisite) and review_invited_identity (the read-only counterpart), so an agent can distinguish it from those siblings. The core verb is somewhat embedded in process narration rather than stated up front, which keeps it short of a 5.

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 sequencing is given: use after redeem_invite and after showing the exact disclosure from review_invited_identity, and only once the invitee gives explicit agreement as a plain-text reply rather than a choice option. It states prerequisites (authenticated active company access plus prior steps) and an explicit confirmation boundary before calling. Nothing about when to invoke is left to inference.

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