Vibe Coding Discoveropensquilla
View on GitHubOpenSquilla — Token-Efficient AI Agent with same budget, higher intelligence density
OpenSquilla is a Python microkernel AI agent runtime: a local model router sends each turn to the cheapest capable LLM, with persistent memory, sandboxed tools, MCP-native integrations and one shared loop across CLI, Web UI and chat channels. Ships desktop/uv installs, skills, scheduling and 20+ provider support.
Use Cases
Run a token-efficient coding/ops agent from CLI or Web UIRoute each turn to the cheapest capable model to cut LLM spendDeploy one agent across Slack, Discord, Telegram, Feishu, DingTalk, QQGive agents persistent memory and local vector searchSchedule autonomous background agent tasks and heartbeatsExecute agent tools inside a layered sandboxConnect MCP servers as native agent toolsUse local models via Ollama for private deployments
Built With
- Language
- Python
- Frameworks
- Starlette · Uvicorn · Pydantic · SQLModel · SQLAlchemy · APScheduler · Typer · Rich · ONNX Runtime · LightGBM · sqlite-vec · Vue · Electron · Docker · uv · Lark/Telegram/DingTalk/QQ bot SDKs
Tags
ai-agent · token-efficiency · model-router · mcp · persistent-memory · multi-channel · local-embeddings · sandbox · cli · web-ui · skills · scheduling · self-hosted · tool-use · llm-router · cost-optimization