# firecrawl/web-agent

🔥 Open-source web data agent optimized for structured web research

Repository: https://github.com/firecrawl/web-agent
Canonical: https://ross.abutalabs.com/products/web-agent
Homepage: https://firecrawl.dev
Language: TypeScript
License: MIT
License Family: permissive
Last push: 2026-04-19T19:49:49+00:00

## Health v2 (maintenance only)
Score: 50/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 78, release rhythm 35, longevity 11
- inputs: {"age_days": 159, "days_push": 136, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1222, forks 160 (observed 2026-08-28T04:04:02.382290+00:00)

## What it is
An open-source TypeScript framework for building autonomous web research agents, layered from a Next.js chat template down to an agent core orchestrator built on LangChain's Deep Agents. It combines Firecrawl's web search, scrape, and interact tools with LLMs in a plan-act-observe loop, supporting skills, subagents, and structured output.

## Use cases
- build an autonomous agent that researches the web and returns structured data
- scrape websites and extract llm-ready markdown or json for my app
- create a chat agent with streaming responses and subagents
- add web search and scraping tools to my vercel ai sdk app
- run deep research on a topic and get a structured report
- scaffold an express api server that powers a web research agent

## When to choose
- you want a ready-made, layered foundation for web research agents in TypeScript
- you already use or like the Firecrawl API for scraping and search
- you need skills, subagents, and structured output out of the box
- you want to fork and customize an agent harness built on LangChain Deep Agents

## When to avoid
- you need a language other than TypeScript/JavaScript for the agent layer
- you want a fully self-contained agent with no dependency on the Firecrawl API or its pricing tiers
- you only need simple one-off scraping without agentic orchestration
- you need a mature, long-proven framework - this project is young and fast-moving

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, web-scraping, rag, llm-inference, search-engine, sdk
- domain: large-language-models, crawlers, web-development, developer-tools
- platform: cross-platform, cli
- tags: web-research-agent, deep-agents, langchain, nextjs-template, express-template, structured-output, subagents, skills, firecrawl, autonomous-agent, ai-agents, nodejs, typescript, web-server

## Member repositories
- firecrawl/web-agent (main) score 50

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:02.382290+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 2026-08-30T06:15:37.506186+00:00, confidence not recorded.
  - readme: https://github.com/firecrawl/web-agent (fetched 2026-08-28T04:04:02.382290+00:00, sha e899f5b11980)
  - homepage: https://firecrawl.dev (fetched 2026-08-29T12:23:58.534997+00:00, sha a807d7d17bcb)
  - site_page: https://docs.firecrawl.dev (fetched 2026-08-29T12:23:58.564293+00:00, sha 8fc2a836e7ce)
  - site_page: https://docs.firecrawl.dev/ai-onboarding (fetched 2026-08-29T12:23:58.566236+00:00, sha 0eb26533db23)
  - site_page: https://www.firecrawl.dev/blog/firecrawl-wikipedia-partnership (fetched 2026-08-29T12:23:58.568442+00:00, sha a649f577b116)
  - site_page: https://www.firecrawl.dev/pricing (fetched 2026-08-29T12:23:58.562410+00:00, sha 20bbcfacfeac)
- Data as of 2026-08-30T08:39:29.467469+00:00.