Save the VM's full machine state to a named snapshot so it can be restored later with vm_start(snapshot=name); requires a VM started with saveable=true.
Save the whole machine state to a named snapshot (takes a minute or more, files are GB). Restore with vm_start(snapshot=name). Only works for a VM started with saveable=true.
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes | ||
| resume | No | ||
| timeout | No |
Changes observed during successful MCP inspections.
v0.2.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations exist, so the description carries the full burden, and it delivers real behavioral facts: the operation takes a minute or more, produces GB-scale artifacts, and requires a VM launched with saveable=true. It does not say whether the VM is paused/resumed during the save or what happens if the precondition fails, so it falls short of fully covering a mutation with zero annotation support.
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 clauses ordered by importance: what it does, the operational cost, the restore route, then the precondition. No filler and every sentence adds information the agent cannot get elsewhere.
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?
Given no annotations, no output schema, and 0% schema coverage, the description covers motivation, cost, and prerequisites well but leaves the resume/timeout parameters and the post-save VM state undefined. Adequate to invoke correctly with the required parameter, incomplete for the optional ones.
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% for 3 parameters, so the description must compensate, but it only glosses 'name' as 'a named snapshot'. Neither the boolean 'resume' nor the integer 'timeout' is mentioned anywhere, leaving two of three parameters documented nowhere an agent can see.
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?
Specific verb ('Save') plus resource ('the whole machine state') and target ('a named snapshot'), which cleanly separates it from vm_snapshots (list) and vm_start (restore path). The parenthetical that files are GB-sized makes the operation's weight unambiguous.
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?
Explicitly names the restore counterpart with the exact call form vm_start(snapshot=name), and states the precondition 'Only works for a VM started with saveable=true'. What's missing is any guidance on when not to save (e.g. running vs stopped state), but the routing context is clear.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.