Use of Hard-coded Credentials Affecting crawl4ai package, versions [,0.8.7)
Threat Intelligence
The probability is the direct output of the EPSS model, and conveys an overall sense of the threat of exploitation in the wild. The percentile measures the EPSS probability relative to all known EPSS scores. Note: This data is updated daily, relying on the latest available EPSS model version. Check out the EPSS documentation for more details.
Do your applications use this vulnerable package?
In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.
Test your applicationsSnyk Learn
Learn about Use of Hard-coded Credentials vulnerabilities in an interactive lesson.
Start learning- Snyk IDSNYK-PYTHON-CRAWL4AI-17661138
- published28 Jun 2026
- disclosed16 Jun 2026
- creditAugust829
Introduced: 16 Jun 2026
NewCVE-2026-56262 (opens in a new tab)How to fix?
Upgrade Crawl4AI to version 0.8.7 or higher.
Overview
Crawl4AI is a 🚀🤖 Crawl4AI: Open-source LLM Friendly Web Crawler & scraper
Affected versions of this package are vulnerable to Use of Hard-coded Credentials via the output_path parameter, which allows arbitrary filesystem paths without validation. An attacker can overwrite or create files on the server by supplying crafted paths. Additionally, the monitor endpoints lack authentication, enabling unauthorized access to sensitive monitoring actions. The /execute_js endpoint permits execution of arbitrary JavaScript in the server's browser, and the JWT signing key is hardcoded, allowing attackers to forge authentication tokens. The monitor dashboard renders unescaped user input, which can be exploited to inject malicious scripts. Furthermore, webhook and crawl endpoints accept unvalidated URLs, enabling requests to internal or cloud metadata services. This is only exploitable if the server is deployed with default or weak JWT secrets, or if network access to the Docker API is not restricted.
Workaround
This vulnerability can be mitigated by setting a strong SECRET_KEY (minimum 32 characters) for JWT, enabling authentication via CRAWL4AI_API_TOKEN, and restricting network access to the Docker API.