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

The annotations already mark this as a non-read-only, non-destructive, non-idempotent, open-world call. The description adds the trait that matters most: this is a consent gate for data leaving the boundary, and the agent must obtain a human answer rather than choose. It doesn't say what happens to the run afterward (resume/terminate), which is the main remaining behavioral gap.

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?

Two short sentences, front-loaded with the condition that triggers use and closed with the operative constraint. Nothing is redundant or padded.

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 2 fully documented parameters, no output schema, and annotations covering safety and idempotency, the description supplies the one thing structured data can't: the human-consent requirement and the data-boundary framing. It would be complete with a note on what the decision does to the run's lifecycle.

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 100% and the schema already defines job_id and the allow/refuse enum, including the instruction that the decision must be the person's. The description reinforces the human-consent framing but adds no new detail about the parameters themselves, so the baseline 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?

The description gives a specific action: answering a paused run's approval prompt, with the scope narrowed to runs that want to send the user's data outside Campaign Pack. That scope is precise enough to distinguish it from mutation siblings like share_campaign or abort_campaign, though it doesn't name any of them explicitly.

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 states the trigger condition well – a run is paused and wants to move data outside the pack – and adds a hard usage rule: 'Ask the person; never decide for them.' What's missing is guidance on the adjacent choice, e.g. that abort_campaign is the way to cancel rather than refuse, or what to do if the person is unreachable.

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