# firecrawl/open-agent-builder

🔥 Visual workflow builder for AI agents powered by Firecrawl - drag-and-drop web scraping pipelines with real-time execution

Repository: https://github.com/firecrawl/open-agent-builder
Canonical: https://ross.abutalabs.com/products/open-agent-builder
Language: TypeScript
License Family: other
Last push: 2025-10-20T15:15:47+00:00

## Health v2 (maintenance only)
Score: 38/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 48, release rhythm 35, longevity 22
- inputs: {"age_days": 321, "days_push": 317, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2618, forks 507 (observed 2026-08-28T04:07:04.663022+00:00)

## What it is
Open Agent Builder is a visual, no-code workflow builder for creating AI agent pipelines powered by Firecrawl web scraping. It provides a drag-and-drop canvas (built with React Flow) for composing multi-step agent workflows with real-time streaming execution, MCP tool integration, and human-in-the-loop approvals.

## Use cases
- build ai agent workflows without code
- visual drag-and-drop web scraping pipelines
- create multi-step ai agent automations
- automated research and content generation workflows
- extract structured data from websites with ai agents
- add human approval steps to automated agent tasks
- run agent workflows with real-time streaming updates
- integrate mcp tools into agent pipelines

## When to choose
- you want to visually design and execute ai agent workflows without writing code
- your pipelines need web scraping or data extraction via firecrawl
- you need human-in-the-loop approvals in automated workflows
- you want mcp protocol support for extensible tool integration
- you prefer a self-hosted next.js app with real-time execution and streaming

## When to avoid
- you need a fully stable production tool - the project is under active development with features still in progress
- you prefer code-first agent frameworks like langgraph directly rather than a visual builder
- you don't want dependencies on external services like firecrawl, convex, or clerk
- you need workflows unrelated to web data, scraping, or llm-driven automation

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, workflow-automation, web-scraping, chat-interface, auth, llm-inference, mcp, rag, developer-tools, gui
- domain: web-development, developer-tools, artificial-intelligence, large-language-models, crawlers
- platform: self-hosted, cross-platform
- tags: visual-workflow-builder, no-code, drag-and-drop, langgraph, convex, clerk, react-flow, nextjs, human-in-the-loop, firecrawl, node-based-editor, template-library, real-time-streaming, mit-license, ai-agents, automation, data-engineering, web-server, nodejs, typescript, docker

## Member repositories
- firecrawl/open-agent-builder (main) score 38

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:04.663022+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:20:22.299582+00:00, confidence not recorded.
  - readme: https://github.com/firecrawl/open-agent-builder (fetched 2026-08-28T04:07:04.663022+00:00, sha 5fcd5c430f33)
- Data as of 2026-08-30T08:39:29.467469+00:00.