Ghost

ghost 👻

[](LICENSE)    [](https://github.com/NousResearch/hermes-agent)  

**A private, unrestricted agentic harness.** A real terminal agent that runs commands, edits files, executes code, and searches the web, with every hosted request routed through OpenGradient's TEE gateway so the model provider never sees your prompts. It answers what you actually ask, drops to a fully-offline local model on demand, and phones home to no one.

Built on the [Hermes Agent](https://github.com/NousResearch/hermes-agent) engine by Nous Research, wired to OpenGradient's gateway and to only open-weight, unrestricted models.

Install (30 seconds)

One deterministic command, no LLM and nothing agentic, installs **and** updates everything (the engine, the privacy stack, the `ghost` commands). uv provisions an isolated Python 3.11 under the hood, so the only prerequisite is `git`.

**macOS, Linux, WSL2:**

curl -fsSL https://raw.githubusercontent.com/OpenGradient/ghost/main/install.sh | bash

**Windows (PowerShell):**

irm https://raw.githubusercontent.com/OpenGradient/ghost/main/install.ps1 | iex

Then:

ghost-login     # connect your account once (browser login)
ghost           # start chatting (default: DeepSeek V4 Pro, private via the TEE gateway)

Re-run the same command, or `ghost update`, to update. From a local clone it's just `./install.sh`. Want the offline local model too? `GHOST_L