CVE-2026-55532 PraisonAI: Origin-validation bypass (startswith prefix match) enables unauthenticated cross-site request forgery against the PraisonAI MCP HTTP server
Attackers can bypass origin validation to perform unauthenticated cross-site request forgery, enabling tool invocation and file writes in PraisonAI.
- Exploited in the wild
- Not known
- Login needed
- Not needed
- Attackable
- Over the internet
- Victim action needed
- A user has to click or open something
- Fix
- Available
What is CVE-2026-55532?
CVE-2026-55532 is a high-severity vulnerability in MervinPraison PraisonAI, published on 25 Aug 2026. An attacker can exploit it over the internet without logging in, and only if a user opens a file or follows a link.
Official description
PraisonAI is a multi-agent teams system. Prior to praisonai 4.6.58, MCP HTTP Stream _validate_origin uses request_origin.startswith(allowed), allowing the attacker-controlled localhost.attacker.com HTTP origin to satisfy the localhost allowlist. A webpage can send Content-Type: text/plain requests without preflight and invoke tools/call without an API key, including file writes that persist agent instructions. This issue is fixed in version 4.6.58.
Which versions are affected by CVE-2026-55532?
3 open source packages ship an affected version, including praisonai and PraisonAI.
| Package | Affected | Fixed in |
|---|---|---|
praisonai PyPI | All versions | 4.6.58 |
praisonai PyPI | All versions | 4.6.58 |
PraisonAI PyPI | All versions | 4.6.58 |
How do I fix CVE-2026-55532?
Upgrade praisonai to 4.6.58 or later.
Is CVE-2026-55532 being exploited?
It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-55532 exists.
Sources
Also known as GHSA-pvph-5j39-v8qc, PYSEC-2026-3892.