firecrawl
AI Agents & AssistantsThe Web Data API for AI - Power AI agents with clean web data
Features
- Search the web and retrieve full page content from results
- Scrape any URL into clean Markdown, HTML, JSON, or screenshots
- Interact with pages via AI prompts or code (click, scroll, fill forms)
- Agent endpoint for automated data gathering with simple descriptions
Security Response History
1 CVE| CVE | Severity | Disclosed | Patched (this tool) | vs Ecosystem Median |
|---|---|---|---|---|
| CVE-2023-4863 KEV |
high
CVSS 8.8
|
2023-09-13 | 2026-02-03 | 2y 5mo / median 2y 4mo |
Recent releases
View all 3 releases →
v2.9.0
Breaking risk
Breaking changes
- OAuth Protected Resource Metadata endpoint path changed to RFC 9728 compliance
- authorization_servers field changed from string to JSON array
Security fixes
- SSRF hardening in Playwright service
- CVE-2026-27606
Notable features
- /interact endpoint for browser automation and dynamic content
- Query format for direct answer extraction without markdown
- PDF parsing modes (fast/auto/ocr) with maxPages control
v2.8.0
Breaking risk
Breaking changes
- Extract endpoint deprecated in favor of /agent endpoint
Security fixes
- CVE-2025-59466 fix
- Lodash prototype pollution vulnerability fix
- Multiple dependency CVE updates
Notable features
- Parallel agents executing thousands of concurrent /agent queries
- Firecrawl CLI and Skill for autonomous agent integration
- Spark model family (Fast/Mini/Pro) with automatic waterfall execution
Weekly OSS security release digest.
The CVE patches and breaking changes that affected production tools this week. One email, every Sunday.
No spam, unsubscribe anytime.
About
Stars
127,747
Forks
7,616
Languages
TypeScript
Python
Rust
Downloads/week
890
↑88%
NPM Maintainers
4
Contributors
154
TypeScript
Types included ✓
