PatchSiren cyber security CVE debrief
CVE-2026-48814 Jovancoding CVE debrief
CVE-2026-48814 is a critical vulnerability in Network-AI, a TypeScript/Node.js multi-agent orchestrator. Versions 5.7.1 and earlier are affected by an issue allowing unauthenticated cross-origin MCP tool invocation. This is due to an empty default secret used by the MCP SSE server. Although CVE-2026-46701 partially addressed this issue in version 5.4.5 by restricting CORS to localhost origins, the empty default secret flaw persisted. The vulnerability enables any non-browser caller, such as curl or SSRF, to invoke all 22 MCP tools without credentials. This issue was fixed in version 5.7.2.
- Vendor
- Jovancoding
- Product
- Network-AI
- CVSS
- CRITICAL 9.1
- CISA KEV
- Not listed in stored evidence
- Original CVE published
- 2026-06-17
- Original CVE updated
- 2026-06-23
- Advisory published
- 2026-06-17
- Advisory updated
- 2026-06-23
Who should care
Organizations using Network-AI versions 5.7.1 or earlier should prioritize patching to version 5.7.2. Security teams and administrators responsible for multi-agent orchestrators and Node.js applications should be aware of this critical vulnerability.
Technical summary
The MCP SSE server in Network-AI versions 5.7.1 and earlier defaults to an empty secret, allowing unauthenticated access. The _isAuthorized() function returns true for empty secrets, and non-loopback binds only produce warnings. This enables unauthorized invocation of MCP tools, including config_set, agent_spawn, and blackboard_write. CVE-2026-46701 partially addressed the issue but did not fix the empty default secret flaw.
Defensive priority
critical
Recommended defensive actions
- Upgrade Network-AI to version 5.7.2 or later
- Review and update MCP SSE server configurations to use secure secrets
- Implement proper authentication and authorization for MCP tool invocations
- Restrict access to Network-AI instances to trusted origins
- Monitor for suspicious activity related to MCP tool invocations
- Consider implementing additional security measures such as IP restrictions and rate limiting
Evidence notes
Information is based on official CVE and NVD sources. The CVE record and NVD detail provide context on the vulnerability. References to GitHub advisories and release notes offer additional insights into the fix.
Official resources
public