Suspicious PraisonAI Unauthenticated Config Endpoint Access via Webserver

PremiumReviewedSigma · Medium · v1
Category
webserver
Author
HuntRule
Published
2026-10-29
Updated
2026-10-29

ATT&CK techniques

Initial Access
  1. Recon

  2. Resource Dev

  3. Execution

  4. Persistence

  5. Priv Esc

  6. Defense Evasion

  7. Cred Access

  8. Discovery

  9. Lateral Movement

  10. Collection

  11. C2

  12. Exfiltration

  13. Impact

What it detects

This rule detects unauthenticated access to PraisonAI agent and MCP configuration endpoints together with version probes exploited through the authentication bypass CVE-2026-44338. Attackers enumerate these paths to harvest agent configuration and confirm vulnerable builds. Access to these endpoints from untrusted sources indicates exploitation of the exposed AI framework.

Related detections9 linkedT1190 — drag to rearrange
Malicious Langflow Build Public Temp RCE Access via Webserver
Suspicious PraisonAI Vulnerability Scanner User Agent via Webserver
Malicious LiteLLM Authentication SQL Injection via Web Request (via webserver)
Possible Marimo Unauthenticated Terminal WebSocket Access via Web Request (via webserver)
Possible React2Shell Next.js Prototype Pollution Exploitation via Web Request (via webserver)
Malicious Marimo Notebook Credential File Access via Shell (via process_creation)
Malicious Ingress-nginx Configuration Injection via Kubernetes Audit (via kubernetes)
Malicious SharePoint ToolShell Post-Exploitation via Encoded PowerShell (via process_creation)
Malicious Web Shell Written to SharePoint LAYOUTS Directory by Worker Process (via file_event)
Suspicious PraisonAI Unauthenticated Config Endpoint Access via Webserver
Pivot detection · T1190 · 9 related

Detection content is published as a reviewed draft. Tune thresholds and exclusions against your own telemetry before enabling this rule for alerting.