mirror of
https://gitee.com/ByteDance/flowgram.ai.git
synced 2025-07-07 17:43:29 +08:00
* feat(runtime): init nodejs runtime * feat(runtime): init folder struct * feat(runtime): interface & test * feat(runtime): basic api & schema interfaces * feat(runtime): init runtime model framework * feat(runtime): create document & node entities * feat(runtime): runtime engine basic execute logic * feat(runtime): node add variable data * refactor(runtime): split to sub domains * test(runtime): document module test * feat(runtime): variable store * feat(runtime): workflow runtime executor * chore(demo): reset initial data * feat(runtime): workflow runtime branch logic * feat(runtime): workflow runtime access to ai model * feat(runtime): workflow runtime data all add to context * feat(runtime): workflow runtime invoke record snaphots * feat(runtime): workflow runtime status * feat(runtime): main api request processing chain * chore(demo): reset initial data * refactor(runtime): types move to interface package * feat(runtime): router access api defines & interfaces * feat(runtime): standardize api register & gen api docs * feat(runtime): create snapshot before node execute * fix(sub-canvas): tips cannot close * chore(demo): reset initial data * feat(demo): make node schema runnable * feat(demo): access test run * feat(runtime): runtime core can run in both browser & server env * fix(runtime): condition value empty issue * feat(runtime): beautify structure data view * feat(demo): test run sidesheet * chore(demo): test run sidesheet button fixed * feat(demo): running node show flowing line * chore(demo): hide node result overflow * chore(demo): reset initial data * feat(runtime): workflow runtime support loop node * fix(container): sub canvas height issue * feat(demo): test run multiple result render * test(runtime): enbale test coverage * refactor(runtime): interface folders structure * refactor(runtime): core folders structure * refactor(runtime): core export apis & access to router * feat(demo): runtime plugin * feat(runtime): server add try-catch protection * fix(runtime): node process reset end time * chore: format json * chore: rush update * refactor(demo): running service move to runtime-plugin as built-in runtime service * fix(runtime): build error * test(runtime): disable nodejs test * fix(demo): test run result key indent width
FlowGram.AI
FlowGram is a node-based flow building engine that helps developers quickly create workflows in either fixed layout or free connection layout modes. It provides a set of interaction best practices and is particularly suitable for visual workflows with clear inputs and outputs.
In the current AI boom, we are also focusing on how to empower workflows with AI capabilities, hence the AI suffix in our name.
📖 Documentation
📦 Packages
| Package | Description | Version |
|---|---|---|
| @flowgram.ai/create-app | App Creator | |
| @flowgram.ai/fixed-layout-editor | Fixed Layout Editor | |
| @flowgram.ai/free-layout-editor | Free Layout Editor |
🎮 Examples
Fixed layout where nodes can be dragged to specified positions, with support for compound nodes like branches and loops.
🚀 Getting Started
# create demo
npx @flowgram.ai/create-app@latest
# in PowerShell
npx "@flowgram.ai/create-app@latest"
# select demo
- fixed-layout # full-feature overview
- free-layout # full-feature overview
- fixed-layout-simple # basic usage
- free-layout-simple # basic usage
🔨 Development
- Install Node.js 18+
nvm install lts/hydrogen
nvm alias default lts/hydrogen # set default node version
nvm use lts/hydrogen
- Clone the repository
git clone git@github.com:bytedance/flowgram.ai.git
- Install required global dependencies
npm i -g pnpm@9.12.0 @microsoft/rush@5.140.0
- Install project dependencies
rush update
- Build the project
rush build
- Run docs or demo
rush dev:docs # docs
rush dev:demo-fixed-layout
rush dev:demo-free-layout
After that, you can start to develop projects inside this repository.
Enjoy it!
Stats
🌟 Contributors
🌟 Contact us
- Issues: Issues
- Lark: Scan the QR code below with Register Feishu to join our FlowGram user group.
Languages
TypeScript
79.9%
MDX
10.9%
JSON-with-Comments
5.3%
JavaScript
2.7%
CSS
0.8%
Other
0.2%

