panic

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

Annotations already declare destructive/non-idempotent/openWorld, but the description adds real state-machine context beyond them: panic latches writes off, cancel is one-shot, resume clears the latch, status is read-only. It does not mention that high-risk modes require the confirmation_token, which the schema covers but the description leaves silent.

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?

Semicolon-delimited clauses with the primary action front-loaded; every clause carries distinct information about a mode and there is no filler.

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 mode-switching control tool with no output schema, the entry points, latch behavior, and reset path are all covered. The main gap is not tying the destructive modes to the confirmation_token gate, which an agent must infer from the schema.

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 50% (the mode enum has no schema description), so the description has to carry mode meaning — and it does, explaining the latch/one-shot/clear/read-only semantics of each value. It adds genuine meaning the enum alone does not.

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+resource ('cancel active background operations') and then disambiguates each of the four modes with distinct behavior. This is enough for an agent to tell it apart from the GUI-automation siblings, though the tool name 'panic' is only one of four modes and the headline verb matches just the 'cancel' case.

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?

The mode clauses imply when each mode applies (resume to clear the latch, status to inspect), but there is no explicit when-to-use/when-not or routing against alternatives such as request_confirmation. Usage must largely be inferred from the enum semantics.

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