This document serves as a collaborative space for researching and discussing vulnerabilities related to Vercel and Next.js deployments.
The primary goal is to:
- Identify and analyze security vulnerabilities disclosed by Vercel on HackerOne
- Understand the attack surface of Vercel-hosted Next.js applications
- Develop detection and exploitation techniques for educational purposes
- Share knowledge and findings with the security research community
- Severity: Critical
- Affected: Next.js React Server Components (RSC)
- Description: RCE vulnerability in Next.js applications using React Server Components
- Status: Known and documented in this repository
- References:
- Technical Analysis: https://react2shell.com/
- Scanner Tool:
CVE-2025-55182.py
- Severity: High
- Affected: Next.js Server Actions
- Description: Source code disclosure vulnerability via Server Actions
- Status: Known and documented in this repository
- Detection: Use
--nextjs-rsc-source-code-disclosureflag in scanner
- Severity: To be determined
- Affected: To be researched
- Description: Additional vulnerability mentioned in scanner banner
- Status: Under investigation
The scanner includes multiple Vercel WAF bypass variants (1-102). Research topics:
- Effectiveness of different bypass techniques
- Detection evasion methods
- WAF signature analysis
- Rate limiting bypass
Key Findings:
- Multiple bypass variants implemented in
waf_bypass_poc.py - Junk data injection for content inspection bypass
- Custom header manipulation
- Timing-based evasion techniques
- Objective: Systematically review Vercel's disclosed reports on HackerOne
- Methodology:
- Identify all publicly disclosed Vercel reports
- Categorize by vulnerability type
- Extract common patterns and root causes
- Develop proof-of-concepts for educational purposes
Resources:
- HackerOne Vercel Program: https://hackerone.com/vercel
- Disclosed Reports: (Contributors please add links below)
Understanding the security model:
- React Server Components (RSC) boundary
- Server Actions security model
- Edge runtime limitations
- Serverless function isolation
Current detection approaches in the scanner:
- Side-channel detection (timing/error-based)
- Direct RCE attempts
- Safe check mode
- Server fingerprinting
- Fork this repository
- Add your findings to this document under the relevant section
- Include proof-of-concept code if applicable
- Submit a pull request with clear description
Use GitHub Issues with the following labels:
research: General research discussionsvulnerability: New vulnerability findingsbypass-technique: WAF/security control bypassestool-improvement: Scanner enhancement ideas
- Only test systems you own or have explicit permission to test
- Follow responsible disclosure practices
- Do not use findings for malicious purposes
- Respect HackerOne's disclosure timelines
- Comply with all applicable laws and regulations
- Next.js Security: https://nextjs.org/docs/app/building-your-application/security
- Vercel Security: https://vercel.com/security
- React Server Components: https://react.dev/reference/rsc/server-components
- HackerOne Platform: https://hackerone.com
- CVE Database: https://cve.mitre.org
- NVD - National Vulnerability Database: https://nvd.nist.gov
- This Repository's Scanner:
CVE-2025-55182.py - WAF Bypass POC:
waf_bypass_poc.py - Burp Suite: https://portswigger.net/burp
- OWASP ZAP: https://www.zaproxy.org
| Area | Status | Contributors | Last Updated |
|---|---|---|---|
| Vercel WAF Analysis | In Progress | - | 2026-01-30 |
| HackerOne Report Review | Not Started | - | 2026-01-30 |
| Edge Runtime Security | Not Started | - | 2026-01-30 |
| New CVE Discovery | In Progress | - | 2026-01-30 |
- Edge Function Sandbox Escape: Research isolation mechanisms in Vercel Edge Functions
- Build-time Injection: Investigate vulnerabilities during the build process
- Environment Variable Exposure: Common misconfigurations in Vercel deployments
- Cache Poisoning: Vercel CDN cache manipulation techniques
- Subdomain Takeover: Orphaned Vercel deployments and DNS issues
- Middleware Bypass: Security controls in Next.js middleware layer
- What other Next.js versions are vulnerable to CVE-2025-55182?
- Are there alternative exploitation methods beyond the current POC?
- How effective are different WAF rules against the bypass techniques?
- What are the indicators of compromise for these vulnerabilities?
- Update Next.js: Upgrade to patched versions
- Input Validation: Implement strict validation on Server Actions
- WAF Rules: Deploy appropriate WAF rules (if not already bypassed)
- Monitoring: Implement detection for exploitation attempts
- Principle of Least Privilege: Limit server-side execution capabilities
- Keep dependencies updated
- Regular security audits
- Implement Content Security Policy (CSP)
- Use security headers
- Monitor for unusual activity
- Initial research document created
- Organized existing vulnerability knowledge
- Outlined research areas and methodologies
- Established contribution guidelines
For private security discussions or responsible disclosure:
- Open an issue with the
researchlabel - Follow responsible disclosure practices
- Share findings that benefit the community
Note: This is an educational repository. All research should be conducted ethically and legally.