CVE-2026-33017

Langflow is a tool for building and deploying AI-powered agents and workflows. In versions prior to 1.9.0, the POST /api/v1/build_public_tmp/{flow_id}/flow endpoint allows building public flows without requiring authentication. When the optional data parameter is supplied, the endpoint uses attacker-controlled flow data (containing arbitrary Python code in node definitions) instead of the stored flow data from the database. This code is passed to exec() with zero sandboxing, resulting in unauthenticated remote code execution. This is distinct from CVE-2025-3248, which fixed /api/v1/validate/code by adding authentication. The build_public_tmp endpoint is designed to be unauthenticated (for public flows) but incorrectly accepts attacker-supplied flow data containing arbitrary executable code. This issue has been fixed in version 1.9.0.

An attacker identifies a public-facing Langflow instance via Shodan or Censys (Langflow exposes recognizable UI fingerprints). They craft a POST request to /api/v1/build_public_tmp/{any_valid_flow_id}/flow with a data payload containing a custom Python node whose code executes a reverse shell via subprocess.Popen — no credentials required. Within seconds they receive a shell on the Langflow server, extract all LLM API keys from environment variables, enumerate connected databases and cloud service configurations, pivot to internal AI infrastructure, and deploy a persistent backdoor. The Sysdig report documents this exact chain occurring in under 20 hours post-CVE disclosure, confirming weaponized exploit availability.