manage_review

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

Annotations already declare destructiveHint=true and readOnlyHint=false; the description goes further by disclosing the human-in-the-loop consent protocol, the phone-only relay constraint for approve/reject, and that get_mode is read-only. It also hints at mutation semantics (editing summary/details/component before approving, conflict resolution via keep=a|b|both, delete requiring confirm_name). It stops short of saying what delete irrevocably removes or what happens to an item's prior state on approve, so not a 5.

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?

Purpose is front-loaded and every clause carries information; there is no filler. The trade-off is extreme telegraphic compression — slashes, semicolons, and parenthetical shorthand ('list (a line each; item_id: one in full)') that costs readability and forces re-parsing of the action list.

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 6-action, 20-parameter mutation tool with no output schema, the description covers the return shapes ({items, total}, the mode, or the result), the consent gate, and the relay fallback. Gaps remain around paging behavior (after/limit) and the reject path's consequences, but nothing critical for correct invocation is missing.

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 description coverage is 100% and the schema already tags each parameter to its action ('approve: new summary', 'reject: the user's reason'), so the baseline is 3. The description reinforces a few relationships — item_id for one-in-full listing, move_to for misfiled items, user_request as the quoted authorization — but adds little beyond what the schema already encodes.

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 opening clause states the resource concretely — 'The user's review queue and mode' — and enumerates what flows through it (summaries, suggestions, imports, clean-ups, deletes, conflicts). The six actions are each named with a short gloss, so an agent can map action values to outcomes. It does not, however, explicitly contrast itself with siblings like manage_todos or manage_problems, which is the main thing keeping this from 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?

It gives explicit triggering conditions per action ('walk: when the user asks what's waiting'), orders prerequisites ('approve: make it a record (edit / move_to first)'), and names an alternative path ('Relay: approve/reject are phone-only by default; share the link'). Most importantly it states hard when-not rules: approve/reject/set_mode require the user to have said so in this chat, quoted in user_request, and it forbids self-approval or mode changes to smuggle in your own writes.

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