Skip to content
KitploitKITPLOIT
ToolsExploitsBlog
Log in
Submit
ToolsExploitsBlog
Submit

Hacking, PenTest, and Cybersecurity Tools for Your Security Arsenal!

Kitploit is a directory of hacking, cybersecurity, and pentesting tools. Discover the latest project updates to find vulnerabilities, analyze systems, automate testing, and strengthen your security.

FeedsContactPrivacy© 2026 Kitploit

Tool Directory

Categories

View all categories
Loading categories
CVE-2026-32913 — I Found a Zero-Day Vulnerability in OpenClaw — Here’s How It Went | Kitploit
Tools/GitHubGitHub/rickidevs/cve-2026-32913
Vulnerability AnalysisWeb SecurityLearning & EducationCurated Resources
GitHubrickidevs/cve-2026-32913

CVE-2026-32913

I Found a Zero-Day Vulnerability in OpenClaw — Here’s How It Went

View Repository
377 months agoNot yet reviewed
Share

While reviewing the source code, I focused on how OpenClaw handles HTTP requests — specifically the fetchWithSsrFGuard() function, which is responsible for making server-side fetch calls.

I noticed something off.

When a request followed a cross-origin redirect (meaning the server sent a 3xx response pointing to a different domain), OpenClaw was supposed to strip sensitive headers before forwarding the request to the new destination. It did — but only for a narrow hardcoded denylist:

Authorization, Proxy-Authorization, Cookie, Cookie2

The problem? That list is incomplete.

Custom authorization headers like X-Api-Key, Private-Token, or any other bearer-style header that developers commonly use — none of those were stripped. They were forwarded as-is to the redirect destination.

This means: if an attacker could control or influence where a redirect pointed, they could receive sensitive credentials that were never intended for them.


Why This Is Dangerous

Imagine your application uses OpenClaw to call an internal API with a custom X-Api-Key header. A malicious server responds with a redirect to an attacker-controlled URL. OpenClaw follows the redirect — and forwards your API key right along with it.

Game over. Your credentials are now in someone else's hands.

CVSS 3.1 Score: 9.3 (Critical)

  • Attack Vector: Network
  • Attack Complexity: Low
  • Confidentiality Impact: High
  • No privileges required, no user interaction needed

The Fix

The maintainers replaced the denylist approach with a safe-header allowlist. Instead of trying to block known bad headers, the new logic only allows known safe headers through on cross-origin redirects — things like content negotiation and cache validators. Everything else gets stripped by default.

This is the correct approach. Denylist-based security is fragile; allowlist-based security is robust.

References

  • CVE Record: CVE-2026–32913
  • GitHub Advisory: GHSA-6mgf-v5j7-45cr
  • Fix Commit: 46715371b0612a6f9114dffd1466941ac476cef5
  • Affected versions: <= 2026.3.2
  • Patched version: >= 2026.3.7

If You're Using OpenClaw

Update immediately to >= 2026.3.7.

If you use custom authorization headers (like X-Api-Key or Private-Token) and you were on an older version, treat those credentials as potentially compromised and rotate them.


Download Tool

Most Popular

View all →

Discover the most used tools by our community.

  • 1

    OsintHunter

    letox4/osinthunter

    🎯 The ultimate OSINT toolkit: investigation graph, confidence scoring, alias families, map, timeline and reports. 49 modules, 400+ sites, no API keys required.

    772
  • 2

    ghostlock-app

    yukonga/ghostlock-app

    GhostLock One-Tap Execution App (CVE-2026-43499)

    505
  • 3

    ARTEX

    baarai/artex

    LLM multi-agent autonomous penetration testing platform with planner/worker agents, asset and exploration graphs, MITM traffic recording, and vulnerability retesting.

    420
  • 4

    redforge

    cybersecurity91/redforge

    Browser-side red team toolkit with 102 offline tools for recon, payload generation, credential attacks, crypto, encoding and engagement reporting.

    345
  • 5

    CyberStrikeAI

    ed1s0nz/cyberstrikeai

    The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next.

    260
  • 6

    antifox

    antifox-ai/antifox

    AI multi-agent autonomous penetration testing platform that runs recon, attack-surface discovery, exploitation validation and evidence-backed report delivery.

    251

Explore all tools

Browse our collection of tools

View all tools →