LangFlow
PulseAugur coverage of LangFlow — every cluster mentioning LangFlow across labs, papers, and developer communities, ranked by signal.
- 2026-10-28 product_launch LangFlow released version 1.9.0 to address a critical command injection vulnerability. source
- 2026-10-24 research_milestone A critical vulnerability in LangFlow allowed arbitrary command execution, detailed in a blog post and advisory. source
- 2026-10-15 controversy A critical authenticated RCE vulnerability (CVE-2026-105740) was disclosed for the AI workflow builder Langflow. source
- 2026-09-09 product_launch LangFlow v1.12.1 was released with bug fixes. source
- 2026-09-02 controversy A critical vulnerability in LangFlow was exploited to steal OpenAI and AWS API keys. source
- 2026-09-01 controversy A critical RCE vulnerability in the Langflow AI framework is being actively exploited by cybercriminals. source
- 2026-09-01 research_milestone A critical vulnerability (CVE-2026-33017) was disclosed for the AI infrastructure platform Langflow, enabling pre-authentication remote code execution. source
- 2026-08-06 controversy A critical vulnerability in IBM's LangFlow platform is being actively exploited, leading to a CISA warning. source
- 2026-07-23 product_launch Langflow released version 1.11.0, including new features and improvements. source
- 2026-07-22 controversy A critical security vulnerability, CVE-2026-0770, was disclosed in LangFlow, enabling remote code execution. source
- 2026-06-11 controversy Langflow experienced a cyberattack that exposed critical vulnerabilities in AI agents. source
11 day(s) with sentiment data
Langflow RCE vulnerability (CVE-2026-105740) actively exploited in the wild
Multiple reports confirm that CVE-2026-105740, a critical RCE vulnerability in Langflow, has been exploited since September 2026. Attackers are attempting to steal sensitive data and access host files, indicating active malicious campaigns targeting unpatched instances of the AI workflow builder.
Langflow to release security-focused update or hotfix within 14 days
Given the critical nature of CVE-2026-105740 and its active exploitation, it is highly probable that the Langflow maintainers will issue a security update or hotfix to address this vulnerability. Users are urged to patch, suggesting a rapid response is expected from the development team.
MCP ecosystem security flaws extend beyond Langflow
The Model Context Protocol (MCP) ecosystem is facing widespread security vulnerabilities, including RCE and SSRF. Langflow is identified as a tool affected by these flaws, but the problem is systemic within the MCP specification and impacts other servers and SDKs, suggesting a broader impact on AI development tools relying on MCP.
AI agent security vulnerabilities are a growing concern, with Langflow being a recent target
The recent ransomware attack exploiting a Langflow vulnerability highlights a critical emerging threat: AI agents themselves becoming vectors for cyberattacks due to their own security flaws. The rapid, autonomous nature of the JadePuffer attack underscores the need for robust security practices in AI agent frameworks.
Langflow to release security patch for CVE-2025-3248 within 7 days
Following the documented ransomware attack exploiting Langflow (CVE-2025-3248), it is highly probable that the Langflow maintainers will prioritize and release a security patch to address this vulnerability. Given the severity and public exposure of the exploit, a rapid response is expected to mitigate further risks and restore user confidence.
-
Company research agent built with LangFlow, Wikipedia, and SEC data
A company research agent was developed using LangFlow, integrating Wikipedia and the SEC EDGAR database via a server called Dataville. This agent can be prompted to research a specific company, such as Walmart Inc., by …
-
Continuous Diffusion Language Models Outperform Speculative Drafters in Generation Tasks
Researchers have compared the performance of Continuous Diffusion Language Models (CDLMs) and speculative drafters, two methods for efficient language generation. The study, using LangFlow for CDLMs and adapting DSpark'…
-
Critical security flaws disclosed for Langflow, vLLM, and Linux · 10 sources tracked
Multiple critical security vulnerabilities have been disclosed for popular AI-related software. Langflow, an open-source tool for building LLM applications, has several critical flaws rated 9.8 by the National Vulnerabi…
-
AI Coding Agent OpenCode Vulnerable to Command Execution via MCP Servers
A vulnerability in OpenCode, an AI coding agent, allowed untrusted MCP servers to execute commands on developer machines via a malicious OAuth sign-in process. This issue, reported by Pillar Security and affecting versi…
-
Langflow vulnerability CVE-2026-105740 allows sensitive data exposure
A security vulnerability, CVE-2026-105740, has been identified in Langflow, an open-source tool for building LLM applications. Exploitation of this vulnerability involves a two-request process where an attacker first wr…
-
LangFlow Patches Critical Command Injection Flaw for Second Time
LangFlow has released a second patch for a critical command injection vulnerability, identified as CVE-2026-105740, affecting versions prior to 1.9.0. This flaw, similar to a previously patched issue (CVE-2026-105697), …
-
LangFlow vulnerability CVE-2026-105740 allows unauthenticated access
A critical security vulnerability, CVE-2026-105740, has been identified in LangFlow, an open-source tool for building and deploying large language model applications. The vulnerability stems from an auto-login feature t…
-
Langflow vulnerability CVE-2026-105740 enables multi-vector attacks
A security vulnerability in Langflow, identified as CVE-2026-105740, allows attackers to gain a shell within the application's container. From this position, attackers can pursue multiple avenues for broader intrusion, …
-
LangFlow vulnerability allowed arbitrary command execution
A critical vulnerability (CVE-2026-105740) in LangFlow, a tool for building AI agents, allowed authenticated users to execute arbitrary bash commands. The flaw, present in versions prior to 1.9.0 and potentially 1.10.3,…
-
New GRAFT method boosts rare word generation in diffusion language models
Researchers have identified a significant limitation in continuous diffusion language models, where they struggle to generate rare words, producing them at only 0.30x the human rate. This issue stems from the models ave…
-
DeepSeek, Alibaba cut AI prices; n8n, LangFlow release security patches
DeepSeek and Alibaba Group have both reduced their pricing for AI services, signaling a competitive market. Concurrently, n8n and LangFlow have issued security patches for their respective platforms, addressing potentia…
-
AI agents weaponized for data theft, platform vulnerabilities exploited · 1 source tracked
The agent layer of AI is emerging as both a powerful tool for attackers and a significant vulnerability. In one instance, an attacker used 51 Claude Code agents to exfiltrate over 628,000 Ukrainian records, demonstratin…
-
Critical RCE vulnerability in Langflow allows arbitrary command execution
A critical remote code execution vulnerability, CVE-2026-105740, has been discovered in Langflow, an open-source platform for building AI applications. The flaw, rated 9.9/10, allows any authenticated user to execute ar…
-
Model Context Protocol Faces Widespread Security Vulnerabilities
Security vulnerabilities have been discovered in the Model Context Protocol (MCP) ecosystem, with multiple critical flaws affecting various servers and SDKs. These issues, including Remote Code Execution (RCE) and Serve…
-
AI Security Woes: MCP Vulnerabilities and Inadequate AIBOMs Highlight Industry Gaps · 3 sources tracked
Security vulnerabilities are increasingly plaguing AI systems, with the Model Context Protocol (MCP) being a recent focus. Multiple critical CVEs have been disclosed, including remote code execution in Langflow and serv…
-
Critical RCE vulnerability found in AI workflow builder Langflow
A critical vulnerability, CVE-2026-105740 with a CVSS score of 9.9, has been identified in Langflow, an open-source tool for building AI and agentic workflows. This authenticated Remote Code Execution (RCE) flaw has see…
-
AI development outpaces security, reviving old vulnerabilities
The rapid advancement of AI development is outpacing the implementation of secure engineering practices. As organizations integrate frameworks such as PyTorch, Hugging Face, and LangFlow into their systems, older vulner…
-
MCP Gateway Centralizes Agent Tool Security
An MCP gateway serves as a centralized security layer for agents interacting with internal MCP servers, addressing the challenge of managing tool access and permissions across multiple agents and teams. Unlike embedding…
-
Critical CVE-2026-93674 in Langflow OSS risks credential compromise
A critical vulnerability, CVE-2026-93674, has been identified in self-hosted Langflow OSS platforms, affecting versions 1.0.0 through 1.12.2. This flaw allows unauthenticated attackers to execute OS commands, potentiall…
-
Critical OS Command Injection Vulnerability Found in Langflow AI Platform
A critical vulnerability, CVE-2026-105697, has been disclosed in the open-source AI agent prototyping platform Langflow. The vulnerability allows for OS command injection, enabling any user to execute arbitrary commands…