Creates a new empty scene in Marmoset Toolbag, discarding unsaved changes without prompting; save large or bake scenes first to avoid crash-related data loss.
New empty scene. Unsaved changes are lost without a prompt. Save a large scene first: Toolbag 5.0.2 has crashed here on big bake scenes.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Changes observed during successful MCP inspections.
v0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Goes well beyond the destructiveHint=true annotation by disclosing that there is no confirmation prompt, that unsaved work is irrecoverably lost, and that Toolbag 5.0.2 has a specific crash risk on large bake scenes. That version-specific hazard is exactly the kind of context structured fields cannot carry.
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 short sentences, front-loaded with the core action and followed immediately by the risk warning. Every sentence carries distinct information; none could be removed without losing a real signal.
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?
Output schema exists so return values need no explanation, annotations cover the safety profile, and there are no parameters. The remaining burden — destructive side effects and a known crash condition — is fully addressed.
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?
Zero parameters, so there is nothing for the description to disambiguate. Baseline 4 applies; nothing is gained or lost here.
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: creates a new empty scene. It's clearly distinguishable from scene_load and scene_save by name and effect, though it never explicitly names those siblings as alternatives.
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?
'Save a large scene first' gives a concrete precondition that implicitly routes the agent to scene_save, and 'Unsaved changes are lost without a prompt' establishes when not to call it casually. It stops short of explicitly naming alternatives or excluding cases like loading an existing file.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.