vm_start

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

With no annotations, the description carries the full burden. It usefully discloses that startup is backgrounded and slow (9-14 min vs 30 s for restore), which is decision-critical. It omits what happens on repeated invocation while running, and what state or handle is returned.

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 what it does and followed by timing plus the required next step. No filler; each clause conveys operational information.

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 no-required-param boot tool with no output schema, the description covers what an agent needs to sequence correctly: it is long-running, backgrounded, and must be followed by safari_ready. Missing only failure/already-running behavior and any return value note.

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%, so every parameter is already documented with defaults. The description only adds meaning to the snapshot path by noting a restore takes about 30 s; it says nothing about cpus, seconds, saveable, or spinlock_timeout_us beyond the schema.

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 and resource: "Boot HackiOS in QEMU in the background." An agent immediately knows this launches the VM rather than managing it. It doesn't explicitly differentiate from siblings like vm_status or wake, though the snapshot-restore alternative is mentioned.

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?

Gives clear sequencing ("Then call safari_ready") and decision-relevant context (9-14 min cold boot vs ~30 s snapshot restore), which tells an agent when to prefer a snapshot. It stops short of stating exclusions, e.g. what to do if the VM is already running.

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