HispasecSeptember 1, 2026🇪🇸Translated from Spanish

Attackers Exploit Critical Langflow and Ruby on Rails Flaws for Credential Theft and C2 Infrastructure

Security researchers have confirmed active exploitation of two critical vulnerabilities: CVE-2026-0768 affecting Langflow and CVE-2026-66066 (nicknamed KindaRails2Shell) in Ruby on Rails. The campaigns follow a clear pattern of initial reconnaissance and secret theft followed by efforts to establish command-and-control infrastructure, raising the risk of eventual remote code execution.

In Langflow, the flaw permits arbitrary Python code execution with root privileges because of inadequate input validation. Observed requests target environment variables such as LANGFLOW_SUPERUSER, patterns matching OPENAI_API keys, and AWS credentials including AWS_ACCESS and AWS_SECRET. Attackers have also attempted to read sensitive paths such as /root/.cache/langflow/secret_key and inspected local artifacts including .ssh directories and the size of .bash_history files.

The Ruby on Rails vulnerability is particularly concerning because it can be chained to achieve remote code execution. CVE-2026-66066 originates from an arbitrary file-read condition that activates when Active Storage processes image uploads using the libvips library. Exploitation requires that the application uses Vips as the variant processor and accepts image uploads from untrusted users. The root cause lies in discrepancies between how Active Storage interprets files and how libvips processes them, potentially leaking secrets used by Rails to protect sessions and encrypt data.

Recoverable secrets include secret_key_base, the Rails master key, database credentials, cloud storage keys, and external service tokens. With these materials an attacker can move from file reads to code execution, pivot across systems, and maintain persistence without valid credentials.

Indicators point to a distributed campaign. Langflow traffic has been linked primarily to Russian origins with confirmed impact on canary instances in the United Kingdom. For Rails, activity has been observed against canaries in Singapore, Israel, and the United Kingdom, with C2 communications directed toward a host in Israel from a single IP located in France.

Recommended defensive actions include identifying publicly exposed Langflow instances and Rails applications that use Active Storage with image uploads. For Rails, the primary mitigation is to update to versions 7.2.3.2, 8.0.5.1, or 8.1.3.1 and review the ruby-vips and libvips chain to block unsafe operations on untrusted content. If patching is delayed, organizations should disable Vips processing or restrict image uploads. Because the focus is on secrets, immediate rotation of secret_key_base, the master key, and any associated cloud, database, or API credentials is advised. Logs should be reviewed for suspicious reads of paths such as /proc/self/environ and config/master.key, while detection rules for C2 traffic should be strengthened.

Related articles

BoletimSecVulnerabilities & Exploits

Click2Shell Flaw in WordPress Core Enables Remote Code Execution via Single Malicious Link

Researchers at pwn.ai have disclosed Click2Shell, a vulnerability in the WordPress core that allows an attacker to install a malicious theme and achieve remote code execution simply by tricking an authenticated administrator into opening a crafted link. The isolated flaw carries a CVSS score of 7.1, but the full attack chain reaches 9.6. The issue stems from an interpretation mismatch between the WordPress.org theme directory and the administrator browser, causing the browser to automatically trigger the install button without any user confirmation or password prompt. Affected versions start from 6.0 and run up to but not including 7.1.1. The vulnerability has been fixed in WordPress 7.1.1 with backported patches released for all supported branches down to version 4.7. No exploitation in the wild had been observed at the time of disclosure, yet the low barrier of convincing an admin to click a link makes prompt patching essential.

Security NEXTVulnerabilities & Exploits

CISA Adds Three Actively Exploited Linux Kernel Vulnerabilities to KEV Catalog

The U.S. Cybersecurity and Infrastructure Security Agency has added three vulnerabilities affecting the Linux Kernel to its Known Exploited Vulnerabilities catalog. The flaws, identified as CVE-2025-39682, CVE-2026-53266, and CVE-2025-39964, are confirmed to be under active exploitation in the wild. CISA is directing all federal agencies to apply patches immediately and to search for indicators of compromise. The vulnerabilities impact the kernel's kTLS TLS processing, the ebtables network bridge component, and the AF_ALG cryptographic interface. Each issue can lead to memory corruption or inconsistent internal state that attackers may leverage for privilege escalation or remote code execution.

HabrVulnerabilities & Exploits

Keycloak Cluster on Three VMs: Engineers Detail Embedded Infinispan Setup, HA Testing Failures, and Bug Fix

A team deployed Keycloak for 25,000–30,000 users across a country using three isolated VMs instead of Kubernetes. They chose Embedded Infinispan with JGroups over an external cluster to reduce failure points and maintain session replication. PostgreSQL with Patroni handled shared storage and JDBC_PING discovery. Testing revealed that full datacenter outages triggered split-brain conditions and 5xx errors that the built-in healthcheck could not resolve. The engineers wrote a Python watcher script to detect multiple coordinators in the JGROUPS_PING table and restart affected containers. They reported the recovery bug to the Keycloak project, which was confirmed and fixed in a later release. The production cluster now survives single-DC loss without session loss.

HabrVulnerabilities & Exploits

From UDP Probe to Working Tunnel: Analyzing Hysteria2 Behavior via QUIC, HTTP/3 and Client Events

A detailed technical study examined how commercial Hysteria2 VPN endpoints respond to different levels of network probing, starting from simple UDP packets and progressing to full QUIC handshakes and authenticated client connections. The research used custom NetProbe tools on Windows and Linux to test real subscription endpoints, capturing both external probe results and decrypted traffic from an instrumented Hysteria2 client. Simple UDP probes received no response and timed out, while properly formed QUIC Initial packets with TLS ClientHello and ALPN h3 successfully completed handshakes using TLS 1.3 and TLS_AES_256_GCM_SHA384. An unauthenticated HTTP/3 GET request triggered immediate connection closure with H3_NO_ERROR (code 0x100), whereas a legitimate client sending POST /auth with Hysteria-Auth headers received the expected 233 status confirming authentication and UDP forwarding support. Subsequent data transfer tests, including parallel requests and file downloads, were observed inside separate QUIC streams after authentication, confirming that the tunnel carried real traffic. The study highlights differences between external probing and authenticated client behavior, providing practical insight into how Hysteria2 servers handle reconnaissance versus legitimate VPN usage.