CVE-2026-55529
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.
Does this matter?
Lower severity and a low EPSS score (0.12%). Track it; it rarely justifies an emergency change on its own.
Description
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 3.1
- 6.9 MEDIUMCVSS:3.1/AV:N/AC:H/PR:N/UI:R/S:C/C:H/I:L/A:N
- EPSS
- 0.12% probability · 2th percentile
- CISA KEV
- Not listed
- Weakness
- CWE-306, CWE-346
- Source
- security-advisories@github.com
References
- https://github.com/MervinPraison/PraisonAI/commit/2f9677abb2ea68eab864ee8b6a828fd0141612e1
- https://github.com/MervinPraison/PraisonAI/releases/tag/v4.6.58
- https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-wj6g-v78p-6fx3
- https://github.com/MervinPraison/PraisonAI/security/advisories/GHSA-wj6g-v78p-6fx3
Source: NVD record, EPSS from FIRST.org, KEV from CISA. Refreshed daily.