Runs scripted browser or device steps through a governed Runner, classifying every action before execution and recording a provenance trail for replay.
Run scripted steps through the governed Runner. driver: 'fake', 'chromium', 'simphone' or 'android' (default: the session body, GHOST_HANDS_DRIVER).
| Name | Required | Description | Default |
|---|---|---|---|
| goal | No | ||
| pages | No | ||
| steps | Yes | ||
| driver | No | ||
| start_url | No | ||
| approve_all | No |
Changes observed during successful MCP inspections.
v0.9.2Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare destructiveHint=true and openWorldHint=true, so the safety profile is partly covered. However, the description does not mention approval requirements, side effects, or what 'governed' entails, and it never addresses the non-idempotent or destructive nature beyond the vague word 'governed'.
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?
The description is short and front-loaded, with no wasted words. It is appropriately sized for what it covers, though it omits critical detail.
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?
This is a complex, destructive, open-world tool with six parameters, nested objects, and no output schema. The description only partially covers the driver parameter and omits the structure of steps, the purpose of goal/pages/start_url, and the meaning of approve_all.
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 0%, so the description carries the full burden. It only explains the driver parameter (with values and a default), leaving goal, pages, steps, start_url, and approve_all completely undocumented.
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 gives a specific verb and resource: 'Run scripted steps through the governed Runner.' It is clear what the tool does, but it does not explicitly differentiate itself from siblings like hands_act or hands_perceive.
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?
There is no explicit guidance on when to use this tool versus hands_act, hands_trail, or hands_perceive. The phrase 'scripted steps' implies multi-step execution, but no alternatives or exclusions are stated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.