CVE-2025-8667
📋 TL;DR
This critical vulnerability in SkyworkAI DeepResearchAgent allows remote attackers to execute arbitrary operating system commands through command injection in the from_code/from_dict/from_mcp functions. The vulnerability affects all versions up to commit 08eb7f8eb9505d0094d75bb97ff7dacc3fa3bbf2, and organizations using this software with internet-facing deployments are at highest risk.
💻 Affected Systems
- SkyworkAI DeepResearchAgent
⚠️ Manual Verification Required
This CVE does not have specific version information in our database, so automatic vulnerability detection cannot determine if your system is affected.
Why? The CVE database entry doesn't specify which versions are vulnerable (no version ranges provided by the vendor/NVD).
🔒 Custom verification scripts are available for registered users. Sign up free to download automated test scripts.
- Review the CVE details at NVD
- Check vendor security advisories for your specific version
- Test if the vulnerability is exploitable in your environment
- Consider updating to the latest version as a precaution
⚠️ Risk & Real-World Impact
Worst Case
Full system compromise allowing attackers to execute arbitrary commands, steal data, install malware, or pivot to other systems in the network.
Likely Case
Unauthorized command execution leading to data exfiltration, system manipulation, or service disruption.
If Mitigated
Limited impact through proper network segmentation, input validation, and least privilege controls.
🎯 Exploit Status
Proof of concept is publicly available on GitHub, making exploitation straightforward for attackers.
🛠️ Fix & Mitigation
✅ Official Fix
Patch Version: Unknown - vendor did not respond to disclosure
Vendor Advisory: None available
Restart Required: No
Instructions:
No official patch available. Consider removing or replacing the software until vendor provides fix.
🔧 Temporary Workarounds
Input Validation and Sanitization
allImplement strict input validation and sanitization for all parameters passed to from_code/from_dict/from_mcp functions
Network Access Restriction
allRestrict network access to DeepResearchAgent instances using firewall rules
🧯 If You Can't Patch
- Isolate affected systems in a segmented network zone with strict egress filtering
- Implement application-level input validation and command whitelisting
🔍 How to Verify
Check if Vulnerable:
Check if your DeepResearchAgent commit hash is 08eb7f8eb9505d0094d75bb97ff7dacc3fa3bbf2 or earlierCheck Version:
git log --oneline -1Verify Fix Applied:
Verify that input validation prevents command injection in from_code/from_dict/from_mcp functions📡 Detection & Monitoring
Log Indicators:
- Unusual command execution patterns
- Suspicious parameters in tool.py function calls
- Unexpected process spawns
Network Indicators:
- Unusual outbound connections from DeepResearchAgent instances
- Command and control traffic patterns
SIEM Query:
Process creation events from DeepResearchAgent with unusual command-line arguments