A critical flaw in the open source AI agent orchestration platform Ruflo, tracked as CVE-2026-59726 and dubbed RufRoot, allowed unauthenticated attackers to execute commands inside the bridge container. Successful exploitation could lead to shell access, API key theft, agent swarm takeover, and poisoning of the AgentDB learning store, and the issue was fixed in Ruflo 3.16.3. #Ruflo #CVE-2026-59726 #RufRoot #AgentDB
Keypoints
- Ruflo is an open source AI agent orchestration platform with agent swarms and MCP tool calling.
- CVE-2026-59726 exposed the POST /mcp endpoint without authentication.
- Default docker-compose settings bound the bridge and MongoDB to all interfaces.
- Attackers could run terminal_execute to gain command execution in the container.
- Ruflo 3.16.3 patched the flaw and included remediation guidance.
Read More: https://www.securityweek.com/critical-ruflo-flaw-lets-attackers-spawn-rogue-ai-swarms/