WiseClaw

WiseClaw is a local-first personal assistant for macOS. It runs a FastAPI backend, supports either a local LM Studio endpoint or the hosted Z.AI API for LLM access, can augment replies with an AnythingLLM-backed second brain, exposes a Telegram bot, and includes a React admin panel for settings, logs, and memory management.

Planned capabilities

Telegram chat with whitelist support
Global model provider switch for Local (LM Studio) or Z.AI
Local LM Studio integration for qwen3-vl-8b-instruct-mlx@5bit
Z.AI integration for glm-4.7 and glm-5
AnythingLLM second-brain context retrieval via workspace chat
Brave or SearXNG-backed web search
Apple Notes integration via AppleScript
File read/write tools
Terminal execution with policy modes
SQLite-backed memory, settings, and audit logs
macOS Keychain for secrets

Repository layout

backend/ FastAPI app and WiseClaw core modules
frontend/ React admin panel
docs/ architecture and rollout notes

Local development

Backend

cd backend
python3.12 -m venv .venv312
source .venv312/bin/activate
pip install .
uvicorn app.main:app --reload

Frontend

cd frontend
npm install
npm run dev

Smoke checks

cd backend
source .venv312/bin/activate
uvicorn app.main:app --reload

Then in another shell:

curl http://127.0.0.1:8000/health
curl http://127.0.0.1:8000/bootstrap
curl http://127.0.0.1:8000/admin/integrations/llm
curl http://127.0.0.1:8000/admin/integrations/telegram

Environment bootstrap

Copy .env.example to .env and fill in only the values you need for the first boot. Secrets such as WISECLAW_ZAI_API_KEY and WISECLAW_ANYTHINGLLM_API_KEY can also be saved later from the admin panel.

1.7 KiB Raw Blame History