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

With annotations only covering the safety profile (readOnly=false, destructive=false, idempotent=false, openWorld=false), the description carries the burden and delivers: queued system invitation email rather than a campaign or access grant, only the intended verified recipient can accept, cancel cannot recall accepted mail, and review decisions require independent authority. These are substantive behavioral traits an agent cannot infer from the annotations or schema.

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 content is dense with useful facts but delivered as one run-on paragraph that opens with a precondition rather than the tool's purpose. Almost every clause earns its place, yet the lack of front-loading and segmentation makes it harder to scan than it needs to be.

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 six-action mutation tool with no output schema and an opaque payload, the description covers confirmation, idempotency/retry discipline, per-action parameters, and the boundary against event administration. It also links a workflow guide and defers delivery-status inspection to separate tooling. Minor gaps remain around a few payload fields and the response shape for uncertain outcomes.

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?

The payload is a free-form object (additionalProperties: {}) that the schema cannot document, so the description must compensate and largely does: inviteId/requestId, email, role values editor/reviewer/publisher/viewer, expectedRevision 0, memberUid/active, requireReview, and reviewId/reviewHash/expectedDecisionRevision/decision values. It omits explanation of payload fields like draft/definition and cursor, so it is not fully exhaustive.

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 names specific operations (inviteCreate, inviteResend, inviteCancel, setMember, setPolicy, decideReview) and the resource they act on (email workspace team/grants), so the umbrella purpose is identifiable. It even differentiates itself from related behavior by contrasting invitation-based membership with setMember's direct verified-account grants. It stops short of 5 because the core purpose is scattered across a dense paragraph rather than stated up front.

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?

It supplies real conditions: explicit organizer confirmation first, show workspace/email/role before confirming, inspect delivery status after an uncertain response, never mint a new ID to retry, and cancel cannot recall accepted mail. It also names an alternative path (setMember manages grants directly without an invitation) and points to the socialloop://guides/email-campaigns resource. It lacks a systematic when-to-use-this-action-vs-that-action statement for setPolicy and decideReview, so it falls short of 5.

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