48 matches found
IBM Langflow OSS configuration error vulnerability
IBM Langflow OSS is an open-source low-code visualization AI workflow building tool developed by IBM. It enables developers to quickly build, deploy, and iterate AI agents and RAG applications through drag-and-drop components. Additionally, it supports deep customization using Python to integrate...
Security Bulletin: Unauthenticated File Upload Vulnerability Allows Disk Space Exhaustion and Path Disclosure in Langflow OSS
Summary Unauthenticated users can upload unlimited files to the Langflow OSS server via the deprecated /api/v1/upload/flowid endpoint without authentication or validation, leading to potential disk space exhaustion DoS and information disclosure through absolute file path leakage in API responses...
Security Bulletin: Path Traversal Vulnerability in File Processing Components Allows Unauthorized File System Access and Potential Remote Code Execution
Summary A path traversal vulnerability exists in multiple Langflow OSS file processing components Docling, Docling Serve, Read File, NVIDIA Retriever Extraction, Video File, and Unstructured API that are based on BaseFileComponent. The vulnerability in the unpackbundle function allows attackers t...
IBM Langflow OSS code injection vulnerability (CNVD-2026-31205)
IBM Langflow OSS is an open-source low-code visualization AI workflow building tool developed by IBM. It enables developers to quickly build, deploy, and iterate AI agents and RAG applications by dragging and dropping components. Additionally, it supports deep customization using Python to...
CVE-2026-14499
CVE-2026-14499 affects Langflow OSS 1.0.0–1.10.1. An authenticated user could execute arbitrary commands with elevated privileges due to improper validation of user-supplied input in the Python Interpreter component, resulting in OS command injection (CWE-78). IBM’s bulletin assigns CVSS v3.1 bas...
CVE-2026-7872 Path Traversal Vulnerability in File Component Leading to Arbitrary File Read and Authentication Bypass
IBM Langflow OSS 1.0.0 through 1.10.0 allows an authenticated attacker to read arbitrary files including the JWT signing key and forge authentication tokens for any user...
EUVD-2026-45282
IBM Langflow OSS 1.0.0 through 1.10.0 allows an authenticated attacker to read arbitrary files including the JWT signing key and forge authentication tokens for any user...
CVE-2026-8635
IBM Langflow OSS 1.0.0 through 1.10.0 allows authenticated users to escalate privileges to superuser by directly manipulating the database, execute arbitrary system commands, and achieve full system compromise with Langflow service permissions...
PT-2026-60834
Name of the Vulnerable Software and Affected Versions IBM Langflow OSS versions 1.0.0 through 1.10.0 Description Improper authentication and access control in the user registration and account provisioning logic allow unauthenticated attackers to create unlimited user accounts. This occurs when...
Security Bulletin: Python Interpreter Sandbox Bypass Leading to Sensitive Data Exposure
Summary The Python Interpreter component and legacy Python REPL Tool component provided flow and agent authors the ability to execute Python code within a scoped globals environment restricted to an operator-configured import allow-list and curated safe builtins. An attacker who could provide inp...
Security Bulletin: Path Traversal Vulnerability in File Component Leading to Arbitrary File Read and Authentication Bypass
Summary A path traversal vulnerability existed in the File component's tar archive extraction functionality. An authenticated user could upload a specially crafted tar archive containing symbolic links to read arbitrary files accessible to the application process. The vulnerability allowed an...
Security Bulletin: MCP Server Configuration Validator Bypass via File Upload API
Summary A validation bypass vulnerability existed in the File Manager API that could allow an authenticated attacker to upload malicious MCP server configurations. The file upload endpoint accepted mcpservers.json files without invoking the MCPServerConfig validator, while the structured MCP API...
Security Bulletin: Unauthenticated User Registration Could Lead to Remote Code Execution
Summary An attacker could create accounts on an exposed instance without authentication through the user registration API. If an operator had configured newly created accounts to be active immediately, the attacker could then log in as that new user and reach code-execution functionality, which...
Security Bulletin: Arbitrary Code Execution in Python Interpreter Component
Summary A vulnerability in the Python Interpreter component allowed authenticated users to execute arbitrary Python code and escalate privileges to superuser level. An attacker with valid credentials could bypass security controls through the PythonREPLComponent, directly access the database, and...
Security Bulletin: Path Traversal in APIRequest Component via Content-Disposition Header
Summary A path traversal vulnerability existed in the APIRequest component when the "Save to File" feature was enabled. The component extracted filenames from server-controlled Content-Disposition headers without proper sanitization, allowing an attacker-controlled HTTP response to write files to...
CVE-2026-7871 Insecure Deserialization in Redis Cache Backend
IBM Langflow OSS 1.0.0 through 1.10.0 allows users with Redis access to execute arbitrary code with full application privileges, compromising all secrets, data, and system integrity...
CVE-2026-10561 Unauthenticated Remote Code Execution in Langflow OSS PythonREPLComponent via Builtins Injection
IBM Langflow OSS 1.0.0 through 1.9.3 has an vulnerability due to an improper isolation of Python execution combined with an authentication bypass that allows an unauthenticated attacker to execute arbitrary code on the host system, resulting in complete compromise...
Security Bulletin: upload filename directly from the multipart Content-Disposition header without sanitization
Summary Langflow OSS 1.2.0 - 1.8.4 are affected by a critical arbitrary file write vulnerability in the files endpoint due to improper handling of uploaded filenames. The application extracts the filename directly from the multipart Content-Disposition header without sanitization and uses unsafe...
