← back

CVE-2026-44338

HIGH · 7.3 EPSS 0.1%
hype LIKELY HACK · 72 hack

Rapid post-disclosure exploitation confirmed; vendor patch released; no PoC code leaked yet.

What: PraisonAI multi-agent framework ships Flask API server with authentication disabled by default (versions 2.5.6–4.6.33), allowing unauthenticated access to /agents and /chat endpoints. CVSS 7.3 HIGH.

Why it matters: Social chatter reports active exploitation via automated scanners within 4 hours of disclosure (2026-05-08). Threat actors observed hijacking agents and draining API quotas. Patch (4.6.34) available. Not yet KEV-listed but rapid weaponization and vendor patching signal real triage pressure.

Where it's seen: Sysdig, HackerNews, security researchers posting exploit timelines and urging immediate patching. Multilingual coverage (English, Japanese) amplifies reach.

RISK: HIGH — Unauthenticated API access; active scanner-based exploitation; API quota theft observed.

Generated by claude-haiku-4-5 from public posts and authoritative metadata. AI can make mistakes — verify against vendor advisories before acting. 5/14/2026, 3:04:36 PM

Description

PraisonAI is a multi-agent teams system. From version 2.5.6 to before version 4.6.34, PraisonAI ships a legacy Flask API server with authentication disabled by default. When that server is used, any caller that can reach it can access /agents and trigger the configured agents.yaml workflow through /chat without providing a token. This issue has been patched in version 4.6.34.

CVSS 3.1 breakdown

Exploitability 3.9 · Impact 3.4
vector CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:L
Attack vector
Network
Complexity
Low
Privileges required
None
User interaction
None
Scope
Unchanged
Confidentiality
Low
Integrity
Low
Availability
Low