agent-harness-install
Install, check, re-onboard and uninstall an AI agent harness (OpenClaw, Deep Agents Code `dcode`, or Hermes) on an Intel Linux host, wired to a local OpenAI-compatible endpoint served by OpenVINO Model Server behind the Intel Inference Router. USE FOR: running the install.sh script, picking a --model and --model-hub, pointing the router at an external endpoint (Ollama, vLLM), Intel Core Ultra iGPU or Arc setups, registering an MCP endpoint, checking status, uninstalling, fixing install failures behind a proxy. DO NOT USE FOR: Windows or macOS hosts, installing Intel GPU drivers or the compute runtime (the installer only detects them), tuning model quality, or general questions about OpenClaw, dcode or Hermes themselves.
Source facts
- Repository
- open-edge-platform/edge-ai-libraries
- Last source activity
- November 9, 2026 at 13:05
- Detected SKILL.md language
- English
- Stars
- 178
- Forks
- 162
Install options
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Review the source files
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Showing SKILL.md
| name | agent-harness-install |
|---|---|
| description | Install, check, re-onboard and uninstall an AI agent harness (OpenClaw, Deep Agents Code `dcode`, or Hermes) on an Intel Linux host, wired to a local OpenAI-compatible endpoint served by OpenVINO Model Server behind the Intel Inference Router. USE FOR: running the install.sh script, picking a --model and --model-hub, pointing the router at an external endpoint (Ollama, vLLM), Intel Core Ultra iGPU or Arc setups, registering an MCP endpoint, checking status, uninstalling, fixing install failures behind a proxy. DO NOT USE FOR: Windows or macOS hosts, installing Intel GPU drivers or the compute runtime (the installer only detects them), tuning model quality, or general questions about OpenClaw, dcode or Hermes themselves. |