Take network VPN peer config

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

Annotations already declare the mutation/destructive/non-idempotent profile, but the description adds substantial context beyond them: one-time consumption with 410 on a second pickup, ciphertext decryptable only by the customer's key, no key creation or decryption, pending-vs-failure semantics via in_progress and reason_kind, and explicit no-auto-retry guidance on transport uncertainty. It also discloses the permission scopes (networking:read vs networking:manage) and backend 2FA/ownership enforcement.

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 one-time nature and mutation semantics are front-loaded in the first two sentences, and most of the length is load-bearing operational detail for an irreversible one-shot action. It repeats the no-auto-retry theme at both the start and the end of the passage, which is minor redundancy in an otherwise tight, dense block.

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 still covers what comes back (a sealed payload to be saved externally), how to poll the returned operation/resource, which flag and reason_kind values indicate waiting versus failure, and the error code for a duplicate pickup. For a three-parameter destructive tool this is complete enough to act on without further reference.

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 coverage is 67%, with peer_id and gateway_id already documented in the schema. The description adds real meaning for acknowledge_one_time_pickup by stressing single consumption and the 410 consequence, which explains why that boolean is mandatory, but it does not restate the UUID format or the exact-ownership requirement in the body text.

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 ('Consume the device's sealed configuration') with an emphatic scope qualifier ('ONCE') that immediately separates it from the many read-only get_network_vpn_* siblings. The opening line 'This is a management mutation despite GET' preempts the most likely misclassification by an agent scanning sibling names.

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 concrete prerequisites and ordering ('Read current capabilities and owned resource state first; obtain approval for the exact change') plus the exact permission scopes required. It stops short of naming the specific sibling tools to call for the capability/state read, so the routing is implied rather than explicit.

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