Active Langflow Exploit Prompts 48-Hour CISA Patch Deadline
Federal agencies have until August 7 to patch critical flaws in IBM Langflow, N-able N-central, and Apache Tomcat amid active exploitation campaigns.
The Cybersecurity and Infrastructure Security Agency (CISA) added four critical vulnerabilities to its Known Exploited Vulnerabilities (KEV) catalog on August 4 and 5, 2026. The mandated updates address active exploitation campaigns targeting IBM Langflow, N-able N-central, and Apache Tomcat. Federal agencies have been ordered to remediate these systems by August 7 or 8, depending on the specific flaw. Threat actors are actively weaponizing these vulnerabilities to execute arbitrary code and hijack administrative accounts.
IBM Langflow Remote Code Execution
IBM Langflow, an open-source visual framework used for building AI agents and retrieval-augmented generation (RAG) pipelines, is facing active attacks following the release of multiple proof-of-concept exploits in late July 2026. Tracked as CVE-2026-9198, the vulnerability carries a maximum CVSS score of 9.8.
Attackers are executing full remote code execution (RCE) by chaining two specific issues within the application. First, an unauthenticated auto-login endpoint issues superuser bearer tokens. Second, attackers use that token to submit malicious payloads to a code validation endpoint designed to execute arbitrary Python code. IBM released a fix for this flaw in Langflow OSS version 1.10.1 in July 2026.
N-able N-central Authentication Bypass
Managed Service Providers (MSPs) using the N-able N-central remote monitoring and management platform face a critical authentication bypass sequence (CVSS 8.2). Threat actors initially exploited a zero-day flaw tracked as CVE-2026-18556. After the initial patch proved incomplete, hackers discovered and began exploiting a bypass, now tracked as CVE-2026-18577.
Compromised N-central servers allow attackers to hijack administrative accounts and pivot directly into managed client endpoints. Attackers are currently leveraging the platform’s “Take Control” feature to deploy persistence mechanisms, primarily routing traffic through malicious Cloudflare tunnels. N-able released emergency hotfix version 2026.3.1.7 on August 2, 2026. As of August 4, security researchers at Huntress reported that 55.6% of internet-reachable N-central cloud servers remained unpatched.
Apache Tomcat and AI-Enabled Exploitation
Apache Tomcat environments using the EncryptInterceptor component in clustered setups are vulnerable to CVE-2026-34486, a high-severity (CVSS 7.5) missing encryption flaw. The Apache Software Foundation fixed this issue in versions 11.0.21, 10.1.54, and 9.0.117.
Researchers at Palo Alto Networks Unit 42 observed active exploitation by a Chinese-speaking threat actor operating under the aliases knaithe or KnYuan. In this campaign, the attacker utilized DeepSeek via the autonomous Hermes agent framework to identify and target internet-exposed devices at scale.
Mandated Remediation Timeline
The CISA directive sets strict deadlines for federal agencies, which serve as a baseline for private sector remediation schedules.
| Vulnerability | Vendor & Product | Action Required | CISA Deadline |
|---|---|---|---|
| CVE-2026-9198 | IBM Langflow | Update to v1.10.1+ | August 7, 2026 |
| CVE-2026-18556 | N-able N-central | Update to v2026.3.1.7 | August 7, 2026 |
| CVE-2026-34486 | Apache Tomcat | Apply vendor updates | August 7, 2026 |
| CVE-2026-63077 | JetBrains TeamCity | Apply vendor updates | August 8, 2026 |
If you operate internet-facing instances of Langflow or N-central, isolate the administrative interfaces from the public internet immediately and apply the specific vendor hotfixes. System administrators must assume compromise for any unpatched N-central instances exposed during the August zero-day window and audit all managed endpoints for unauthorized persistent tunnels.
Get Insanely Good at AI
The book for developers who want to understand how AI actually works. LLMs, prompt engineering, RAG, AI agents, and production systems.
Keep Reading
How to Use Symbolic Execution for Automated BPF Analysis
Learn how Cloudflare uses the Z3 theorem prover to instantly generate magic packets and reverse-engineer BPF bytecode for security research.
Active RCE Exploits Target 7,000 Exposed Langflow Instances
Attackers are actively exploiting a path traversal vulnerability in Langflow's file upload endpoint to achieve unauthenticated remote code execution.
Hackers Exploit Critical Flowise RCE Bug With 10.0 CVSS Score
A maximum-severity code injection flaw in Flowise is under active attack, putting thousands of exposed AI application instances at risk of full system takeover.
Pangram 4 Replaces Watermarks With Style Analysis in $9M Round
Pangram secured $9 million to launch AI detection models that identify machine-generated text and images without relying on metadata or watermarks.
OSAA Forms With $250M to Standardize Open AI Security Tools
Nvidia, Microsoft, and Intel have launched the Open Secure AI Alliance to build standardized, hardware-agnostic defensive tools against weaponized AI models.