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

Annotations indicate readOnlyHint=false, openWorldHint=true, idempotentHint=true, destructiveHint=true. The description adds context: it explicitly states that refusal of mandatory items remains blockers, that this does not approve final content, and gives the explicit prohibition on accessing the mailbox. This clarifies the non-readonly and destructive nature (it sends something external) and the idempotency implications (use exact idempotency key) though the description does not restate the idempotency key behavior because that's in the schema. It adds value beyond annotations by detailing what the action does not accomplish.

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 description is concise, three sentences, with the primary action in the first sentence, constraints and exclusions in the next two. No filler, each sentence earns its place. Front-loaded with the main purpose.

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?

The tool has 3 parameters, all required, with nested objects, and no output schema. The description explains the overall purpose and key constraints (decision exactly once, blockers, no mailbox access). It does not describe the return value (e.g., confirmation status or next steps), but given the absence of an output schema, it could have provided a brief note. However, the existence of sibling tools like `helvabase_confirm_plan_agreement` suggests the workflow, and the description clearly states the action and its boundaries. Minor gap: no mention of what happens after sending (e.g., waits for confirmation), but it is acceptable.

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 33% (only idempotencyKey has a description). The description explains the high-level role of 'revision' and 'selection' but does not detail each field beyond what schema properties indicate. The schema already defines `outputJobId`, `payloadHash`, `acceptedItemIds`, `rejectedItemIds`, and `idempotencyKey` with patterns. The description adds the constraint that 'Every production item must be decided exactly once', which gives context to the selection arrays, and mentions 'revision' implicitly. It does not significantly add new semantics beyond the schema; baseline 3 is appropriate.

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 states a specific verb ('Send'), a resource ('plan and...selection to...email'), and the purpose ('for agreement'). It distinguishes itself from siblings like helvabase_request_review or helvabase_confirm_plan_agreement by making explicit it is about sending the plan for agreement, not confirming. The phrase 'This does not approve final content' clarifies scope, separating it from confirmation tools.

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?

Explicitly states when to use: 'Send the complete exact plan and accepted/rejected production selection to the user's verified email for agreement.' It also provides exclusions: 'Never access the mailbox to retrieve the confirmation code' and 'Refused mandatory items remain blockers', guiding the agent on constraints. However, it doesn't name an alternative tool for when not to use it, but the context of siblings and the specific action makes alternatives implicit.

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