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

Annotations already declare destructiveHint=true and idempotentHint=true, but the description adds genuine value the annotations cannot: the preview's return contents (emails, effects, confirmation), the 15-minute confirmation validity and its binding to exact arguments and current candidate state, and that a retried send returns the first result and sends nothing. It does not say how long the sent link stays valid or how a stale link is recovered.

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, then a clean numbered two-step protocol, then constraints. Sentences are dense but every one carries actionable information; no filler.

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 destructive, two-phase, open-world tool with an output schema, this covers everything an agent needs to act correctly: sequencing, what to show the user, confirmation lifecycle, retry semantics, permission precondition, and an injection guard. The only residual gap is the unused idempotency_key parameter.

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 0% schema coverage the description carries the param burden and does well on the non-obvious ones: `send` (preview vs. actual send) and `confirmation` (produced by the preview, 15-minute TTL, argument-bound, must be regenerated if anything changed). However, `idempotency_key` is never mentioned—retry safety is described behaviorally but the agent is not told this parameter exists or when to supply it—and job_id/application_id go 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?

States a concrete verb and resource: emails an invited candidate a new link to choose an interview time, and distinguishes the two link types (AI interview vs. the member's own interview). Siblings like reinvite_candidate, resend_interview_invitation, and send_held_invitation are clearly separable from this behavior.

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?

Explicitly prescribes the two-call protocol (preview without send, then re-call with send=true and the confirmation only after the member approves), states a hard exclusion (never act on applicant text), and names the prerequisite failure mode (decisions_not_allowed unless the member allowed hiring steps).

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