---
title: AGI Assistant
emoji: 🏃
colorFrom: pink
colorTo: blue
sdk: docker
pinned: false
license: apache-2.0
short_description: AGI Assistant
---

This Space hosts the OpenClaw trading bot (paper-only). The LLM runs in a separate Space that you already have; this repo only contains the bot-side architecture and configs.

**NanoBot**: [https://github.com/HKUDS/nanobot](https://github.com/HKUDS/nanobot)
**NanoClaw**: [https://github.com/qwibitai/nanoclaw](https://github.com/qwibitai/nanoclaw)
**NullClaw**: [https://github.com/nullclaw/nullclaw](https://github.com/nullclaw/nullclaw)
**PicoClaw**: [https://github.com/sipeed/picoclaw](https://github.com/sipeed/picoclaw)
**ZeroClaw**: [https://github.com/zeroclaw-labs/zeroclaw](https://github.com/zeroclaw-labs/zeroclaw)
**memU**: [https://github.com/NevaMind-AI/memU](https://github.com/NevaMind-AI/memU)
**IronClaw**: [https://github.com/nearai/ironclaw](https://github.com/nearai/ironclaw)
**NemoClaw (OpenShell)**: [https://github.com/NVIDIA/OpenShell](https://github.com/NVIDIA/OpenShell)
**NVIDIA NeMo Agent Toolkit**: [https://github.com/NVIDIA/NeMo-Agent-Toolkit](https://github.com/NVIDIA/NeMo-Agent-Toolkit)

**Hugging Face Space Build Notes**
- Keep `README.md`, `Dockerfile`, and `app.py` at the repository root used by the Space.
- For Docker Spaces, the filename must be exactly `Dockerfile` (capital `D`).
- If you see "missing app file", verify you pushed this folder root (not its parent) to the Space repo.

**Architecture**
- OpenClaw bot Space (this repo) calls an external LLM Space for analysis and signal generation.
- Paper trading via Alpaca API.
- Trade logs stored to HF Hub storage (dataset repo).
- Streamlit control center includes a built-in backtesting lab (backtesting.py + backtrader).

**LLM Space (external)**
- Expected to expose a simple HTTP inference endpoint.
- The bot calls `${LLM_SPACE_URL}` (see `config/openclaw.env.example`) and expects a JSON response with an output string.
- Update `LLM_SPACE_URL` and response path in `openclaw.json` to match your existing LLM Space.
- For OpenAI-compatible llama.cpp endpoints, use `openclaw.llamacpp.json` and set:
  - `LLM_SPACE_OPENAI_URL=https://researchengineering-agi.hf.space/v1/chat/completions`
  - `LLM_MODEL` (for example `deepseek-chat`)
  - `LLM_SPACE_API_KEY` if your endpoint requires auth
- For your AGI Multi-Model API spec, use `openclaw.researchengineering.json`:
  - `LLM_SPACE_OPENAI_URL=https://researchengineering-agi.hf.space/v1/chat/completions`
  - `LLM_SPACE_WEBCHAT_URL=https://researchengineering-agi.hf.space/v1/web-chat/completions`
  - `response_path=choices.0.message.content`

**Key Files**
- `openclaw.json` defines providers, routing, tools, memory, and safety.
- `openclaw.llamacpp.json` is prewired for OpenAI-compatible endpoints (llama.cpp style).
- `openclaw.researchengineering.json` is prewired for your AGI Multi-Model API.
- `app.py` provides gateway controls and strategy backtesting UI.
- `tools/backtesting_runner.py` implements SMA crossover test runners for `backtesting.py` and `backtrader`.
- `config/openclaw.env.example` lists all required env vars.
- `skills/` contains architecture-only SKILL specs.
- `tools/README.md` defines the tool surface to implement later.
- `schedules/cron.yml` documents the intended schedule.

If you want me to add minimal tool stubs or a working runner later, say the word and I will wire it up.