#6969th of 80 in Agent platforms
refly
Builder for versioned agent skills, exported as REST APIs, webhooks or MCP
- Stars
- 7.6k
- License
- custom license
- Last commit
- Nov 2026
- Last release
- Feb 2026
- Language
- TypeScript
7,122 on 21 Apr 20267,553 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Refly is a workflow builder where you describe a process in natural language (Vibe Mode) or wire nodes on a canvas, then save it as a versioned skill. Skills can be called through a REST API, triggered by webhooks (Lark/Feishu), or installed into Claude Code via the refly CLI. It supports MCP servers and lists 3,000+ tool integrations, with a hosted workspace and a self-deployment guide.
Who it is for: Teams turning SOPs into reusable agent skills callable via API
Strengths
- Workflows publish as REST endpoints, webhooks, or Claude Code skills via the refly CLI
- Runtime can be paused and re-steered mid-run, with repair of failing workflows
- Central skill registry with versioning, sharing and audit logs for teams
- Supports MCP servers and a tool/provider catalog in config/provider-catalog.json
Weaknesses
- Custom ReflyAI license: Apache 2.0 plus extra restrictions, not OSI-standard
- Cursor export is marked coming soon; the Lark/Feishu guide is unfinished
- No Docker or compose files detected in the repo; deployment is only in linked docs
- Claims such as 100% reliability and 3-minute builds come with no benchmarks
What it needs
- Compose
- Needs LLM provider API key (e.g. OpenAI, Anthropic)
- Models: OpenAI, Anthropic, providers in config/provider-catalog.json
- port 5700
Quick answers
Is refly open source?
Its code is public on GitHub under its own license terms; read them before you use it at work.
Can you run refly with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Also in Agent platforms
See all 80| Rank | Project | Score |
|---|---|---|
| 1 | AutoGPTRole-based AI agents coordinated by one lead agent, plus a visual workflow builder | 95 out of 100 |
| 2 | daily_stock_analysisLLM-driven daily analysis of watchlist stocks across six markets, pushed to chat apps | 95 out of 100 |
| 3 | LangflowVisual builder for AI agents and workflows, exposed as API or MCP server | 94 out of 100 |
| #4 | deer-flowAgent harness running sub-agents, memory and sandboxes with pluggable skills | 93 out of 100 |
| #5 | DifyVisual builder for LLM workflows, RAG pipelines and agents, with APIs | 92 out of 100 |