dimos

Simulates A* route planning and MuJoCo dynamics for a 2D office rover in the DimOS workspace.

465|41|Updated Aug 4, 2026
One-click install
npx skills add https://github.com/autonomous-ai/openharness --skill dimos-autonomous-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dimos
Source: https://github.com/autonomous-ai/openharness/tree/main/store/agents/dimos/skills/dimos
Command: npx skills add https://github.com/autonomous-ai/openharness --skill dimos-autonomous-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? It lets you create and inspect robotics simulation projects in the DimOS Harness workspace, running mission planning and physics simulation locally without needing physical robot hardware. ## Core Features & Use Cases - Mission Simulation: Run A* route planning with real MuJoCo dynamics on a two-dimensional office rover model via the studio toolchain. - Result Inspection: Review run artifacts in out/runs/<id>/, including mission.json, mission.xml, arrival status, travel time, distance, and wall contacts. - Optional Upstream Integration: Read DimOS daemon status and consult upstream quickstart and installation docs before working with the full DimOS stack. - Use Case: Pick an open destination in the office map, run the simulate command, then inspect the recorded MuJoCo positions to verify the rover arrived without wall contacts. ## Quick Start Ask the agent to run a simulation to a chosen destination in the office map and report whether the rover arrived, its travel time, and any wall contacts.

Frequently Asked Questions about dimos

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a robot mission simulation in the DimOS workspace?

Run the simulate command through the studio toolchain's run.sh script after choosing an open destination in the office or gallery map. Results are written to out/runs/<id>/ with mission.json and mission.xml for inspection.

How do I inspect simulation results and robot trajectories?

Check out/latest.json for the current run, then read mission.json and mission.xml in the run directory. Verify arrival, final position, travel time, distance, and wall contacts; replay uses the recorded MuJoCo positions.

Does the DimOS starter simulate a real Unitree robot?

No, the local starter is an original two-dimensional planar office rover using MuJoCo dynamics, not a Unitree robot or physical execution. Simulation success never implies a physical robot reached its destination.

What happens when a simulation destination is inside an obstacle?

A destination inside the inflated obstacle boundary fails explicitly and the failure is recorded in the verdict. The last successful run is preserved, and you should choose another destination instead of inventing a path.

Can I use the full DimOS daemon and perception stack with this workspace?

Yes, but it is an optional larger installation fetched at a pinned source commit. Read the upstream quickstart and installation docs first; the local dimos action only reads daemon status.