About this project

FlowGram is a workflow development framework designed to help developers build custom AI workflow platforms. It provides a suite of built-in tools including a visual canvas (supporting both free and fixed layouts), a form engine for node configuration, a variable management system, and pre-built materials such as LLM, Condition, and Code Editor nodes. The framework enables visual composition of workflows where nodes can be arranged and connected. The free layout canvas allows arbitrary node placement and free-form connections, while the fixed layout supports structured arrangements with compound nodes for branches and loops. The form engine handles data management for node configurations, including validation and side effects. The variable system manages data flow with scope constraints and type inference. FlowGram is distributed as a toolkit; developers integrate it to create their own workflow platforms. The documentation covers setup, core components, and advanced usage. It is used in projects like Coze Studio for AI agent development, NNDeploy for AI model deployment, and Certimate for SSL certificate management.