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

Annotations already declare readOnlyHint/destructiveHint false and openWorldHint true, yet the description adds genuinely new behavior: that answers are persisted onto the request and into the audit trail, that access/portability records become a file attachment, that the call is idempotent-unsafe in effect (re-running re-queries vendors), and which roles may invoke it. It also narrows the read-only claim to the vendor side, which is the important nuance for a caller — though that local write sits in mild tension with readOnlyHint and is disclosed rather than hidden.

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 lead sentence front-loads the action and the side effect, then scope, then return shape, then safety and preconditions — a sensible order. It is dense but nearly every clause carries information; the duplicated task_id scoping sentence is the only mild redundancy.

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?

With no output schema, the description takes on the return-value burden and does so ('Returns per system: found, how many records, the vendor's summary and reference'), plus file-saving behavior for access/portability requests. Combined with stated preconditions and vendor-side read-only guarantees, an agent has enough to call it correctly; only pagination/volume or failure-when-no-connection behavior is left unstated.

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 each parameter is already documented (site slug/id/name, request id prefix, task_id scoping including 'runs even if the task is already done'). The description restates the task_id scoping rule but adds no format, edge-case, or interaction detail beyond the schema, so the baseline of 3 applies.

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 specific verb and resource: query the site's connected apps for the request's requester via each vendor's API, then record answers on the request and audit trail. This is clearly distinguishable from siblings like list_connections (enumerate connections) and get_request (read the request), because it explicitly reaches out through vendor APIs and writes results back.

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?

Gives a clear scope rule for when to pass task_id ('Only this task's system… Omit for every open task with a connection') and states preconditions ('The request must be verified and open; needs the owner or staff role'), which tells the agent when the call will fail. It does not name an alternative tool for the case where no connection exists (e.g., list_connections first), so it stops short of a full when/when-not statement.

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