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

Annotations already declare destructiveHint=true, readOnlyHint=false, openWorldHint=true and idempotentHint=false, so the safety profile is covered. The description adds genuinely new behavior: the job is a persistent systemd unit that outlives the call and agent updates, it returns a job_id, and destructive invocations go through a user approval gate. It does not mention that repeating the call creates a second job or that the job may need stopping, but the added context is substantial.

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?

Three tight sentences, front-loaded with what the tool does and the fact that the job persists, then the return value and the follow-up tool, then the approval caveat. No filler and nothing redundant.

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 correctly supplies the return value (job_id) and names the tool for reading results, and it covers the approval workflow implied by the destructive annotation. Minor gaps remain: no mention of `stop_job`/`list_jobs` for lifecycle management or of what happens when a job fails.

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 83%, so the schema already explains cwd, why, name, server and approval_id; the description adds no syntax, format or default detail for any of them. The baseline 3 applies since the schema carries the parameter burden and the description only implies the approval flow that the `why` and `approval_id` fields serve.

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 and resource ('Start a long command ... in the background') and adds crucial scope: it runs as the invoking user, in its own systemd unit, and survives the call returning and agent updates. That distinguishes it from foreground `run`, from `job_output` (reading progress), and from `list_jobs`/`stop_job` without needing the schemas.

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 gives a clear usage context via concrete examples (installs, composer, compiles, imports, backups) and routes the agent forward to `job_output` for progress. It also flags the approval requirement for destructive commands 'like run', implicitly contrasting with `run`. It never explicitly says when NOT to use it (e.g., short commands should use `run`), so the routing is strong but not complete.

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