Control an owned local job

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

Annotations already declare destructiveHint=true and non-idempotent, and the description adds real context on top: Stop is terminal ('Never restart after a user Stop') and the tool refuses arbitrary watch expressions or shell commands. What it still omits is the effect of a stop on job state/progress and any authorization requirements, which keeps it below a 5.

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 tight sentences, front-loaded with the available operations, followed by a safety constraint and a capability boundary. Every clause carries weight; only the negative 'No arbitrary watch expressions' sentence could be considered secondary.

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 destructive, non-idempotent control tool with 6 undocumented parameters and no output schema, the description is too thin — it never explains what the opaque target/sourceRevision/job hashes are, what mode/points require, or what a successful control action returns. The agent can guess the intent but not call it correctly on the first try.

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 0%, so the description carries the full burden, and it only partly does: 'Stop/Step/Continue/set breakpoints' maps onto the operation enum and hints that points are breakpoint coordinates, with 'Step' loosely implying the into/over/out mode. It says nothing about the required target, sourceRevision, or job identifiers, leaving 3 of 6 parameters — including 3 required ones — completely unexplained.

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?

Names specific verbs (Stop, Step, Continue, set breakpoints) against a specific resource (this grant's matching job / its paused Debug session), so the agent knows exactly which actions it exposes. It does not explicitly distinguish itself from close siblings like crabrix_runtime_start, crabrix_runtime_job, or crabrix_runtime_inspect, which keeps it from a 5.

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 'paused Debug session' phrasing implicitly signals that step/continue only apply in a debug context, and 'Never restart after a user Stop' gives one hard usage constraint. But there is no explicit when-to-use-this-vs-sibling guidance (e.g., how it differs from crabrix_runtime_job or crabrix_runtime_inspect), leaving the agent to infer routing.

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