CVE-2026-55529

PraisonAI is a multi-agent teams system. Prior to praisonai 4.6.58, the MCP HTTP Stream _validate_origin method accepts request_origin.startswith(allowed), so the attacker-controlled localhost.evil.example HTTP origin matches the localhost allowlist. Without an API key, a malicious webpage can submit tools/call requests to the local MCP server and execute exposed tools. This issue is fixed in version 4.6.58.

CVSS v3.1

6.9 MEDIUMCVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:L/A:N

CVSS v3.1

6.9 MEDIUMCVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:L/A:N

EPSS

0.12% chance of exploitation in the next 30 days, 2th percentile

Published

2026-08-25

Updated

2026-08-27

Proof-of-concept exploits (1)

• advisories/GHSA-wj6g-v78p-6fx3

References

• CVE Record
• NVD entry

Related

• CVE-2026-55528
• CVE-2026-5553
• CVE-2026-55527
• CVE-2026-55530
• CVE-2026-55526
• CVE-2026-55531