FastClaw
A lightweight Python AI agent framework built on event-driven, graph-based orchestration, with multi-channel messaging, cron scheduling, and streaming output.
FastClaw is a lightweight AI agent framework in Python built around event-driven agent orchestration: it represents agent behavior as a directed graph with conditional branching between nodes, such as tool execution and response generation, rather than a single linear script. Responses stream in real time as the underlying language model generates them, and a persistent, multi-session conversation history is kept across restarts.
An extensible tool system runs shell commands, skills, and custom integrations, and a built-in cron-style scheduler runs automated tasks on a defined schedule. It ships with Feishu and iMessage channel integrations, and can call into Python's data ecosystem, including libraries such as NumPy and pandas, directly from agent tools. It installs from PyPI or by cloning the repository, defaults to a DeepSeek chat model, and configuration, including which model to use, can be changed either from a local web interface or by editing a JSON file directly.