Mahal Systems MobileWorld
Fits mobile agents in shopping and workplace apps. Borrow the backend-database checks for state and the user-interaction tasks.
What it asks the agent to do
- complete a workflow across several apps (chat, social, e-commerce)
- ask the user a clarifying question before acting
- combine GUI steps with MCP tool calls
What to borrow from it
sandbox designgrading designuser simulator
Caveats
New and single-vendor; apps are open-source stand-ins, not the real products; best framework 51.7 percent, best end-to-end model 20.9.
Details
| Workloads | mobile, shopping, workflow automation, tool calling |
|---|---|
| Interaction | multi turn simulated user |
| Environment | mobile |
| End state checked | Yes |
| Tools | Android emulator snapshot, self hosted Mattermost / Mastodon / Mall4Uni backends, MCP tools |
| Grading | state diff, exact match, programmatic check |
| Size | 201 |
| Data | synthetic |
| Personal data | none |
| License | Apache-2.0 |
| Harness | own repo |
| Status | active |