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

The annotations (readOnlyHint=true, idempotent, non-destructive) align with 'Preview writes/sends nothing', and the description adds genuinely non-redundant behavior: the two-phase human-approval gate, required scopes (team:invite vs team:write), stale-card handling, and a secret-exposure rule about tokens and login links. The one wrinkle is that a readOnly tool is said to 'need team:write', which blurs whether those scopes are validated here or consumed at execution; it is ambiguous rather than contradictory.

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?

Dense and front-loaded with no filler; nearly every clause carries a rule an agent could otherwise get wrong. The shorthand ('ambiguous delivery require read without retry', 'existing membership/invite identity or newInviteId') is terse to the point of being cryptic, which slightly hurts parsing.

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 nested oneOf change object with no schema descriptions and no output schema, the definition covers the hardest parts: where expectedRevision comes from, exact-recipient/role validation, owner-protection rules, and the approval gate. It stops short of enumerating the four change types and does not say what the preview returns, so the agent still has to open the schema to construct a valid payload.

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 description coverage, the description must carry the parameter load, and it does explain the most important one: expectedRevision must be copied exactly from get_team_access, implying optimistic-concurrency semantics the schema cannot convey. It also names recipient/role and newInviteId, but says nothing about the access/revoke_invite variants, the active flag, locale, or the allowed role enums, leaving most of the oneOf shape 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?

The description frames a concrete prepare/validate step for team membership and invite changes, and explicitly says 'Preview writes/sends nothing' and 'Only the human show_team_approval card can execute', so the agent can tell it is a non-executing preparation tool. It does not, however, distinguish itself from the very close sibling prepare_team_access_change, nor does it open with a plain statement of what it prepares.

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?

Strong when-to-use signals: 'After explicit request', 'Only the human show_team_approval card can execute', and 'Expired/stale cards and ambiguous delivery require read without retry' all bound when the tool may be called and what to do on a failed precondition. What is missing is explicit routing against alternatives such as prepare_team_access_change or confirm_team_change.

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