CVE-2026-102697

Ollama versions 0.14.0 before 0.31.2 are vulnerable to an incorrect authorization flaw within the experimental agent mode Bash tool approval mechanism . The root cause is a failure to properly parse shell syntax, which allows attackers to bypass session approval requirements. By using prompt inje...

Meta’s Muse sent a Facebook Marketplace buyer to a seller’s home

A Facebook Marketplace buyer arrived at a seller’s apartment to collect a keyboard. The seller wasn’t home and didn’t know anyone was coming. Meta’s AI assistant Muse had handled the conversation, shared his address, and arranged the visit without telling him. We often write about AI misalignment...

The Blue Agent POV: Investigating Multi-Platform Data Exfiltration Across AWS and GitHub

See how the Blue Agent investigated a multi-platform attack in minutes, following evidence across AWS and GitHub to uncover compromised credentials, stolen source code, and custom data exfiltration tooling...

Security update for amazon-cloudwatch-agent

This update for amazon-cloudwatch-agent fixes the following issues: CVE-2026-84303: google.golang.org/grpc: xDS RBAC HTTP filter implementation issue allows for bypass of authorization policies via mixed-case or canonical-case header matches bsc1279334. CVE-2026-84304: google.golang.org/grpc: hea...

SUSE-SU-2026:4388-1 Security update for amazon-cloudwatch-agent

This update for amazon-cloudwatch-agent fixes the following issues: - CVE-2026-84303: google.golang.org/grpc: xDS RBAC HTTP filter implementation issue allows for bypass of authorization policies via mixed-case or canonical-case header matches bsc1279334. - CVE-2026-84304: google.golang.org/grpc:...

PT-2026-102946

Name of the Vulnerable Software and Affected Versions Fireware OS affected versions not specified Description A path traversal issue in the WebUI management agent allows an authenticated administrator to read or list arbitrary files on the local filesystem by sending a specially crafted managemen...

Concealing LLM-Based Multi-Agent Topology Via Phantom Structure Injection

Driven by the rapid advancement of large language models LLMs, LLM-based multi-agent systems MAS have emerged as a powerful paradigm for collaborative reasoning over complex tasks. A key design element of MAS is the communication topology, which governs information flow among agents and often...

SKILLLITE: Evidence-Guided Malicious Skill Auditing with Compact LLMs

As LLM-based agents perform increasingly complex tasks, Agent Skills have emerged as a flexible mechanism for extending their capabilities. An Agent Skill packages task-specific instructions with executable components and auxiliary resources to provide specialized functionalities. However, the...

One Pipeline Does Not Fit All: TAILOR, a Type- and State-Aware Framework for CVE Reproduction

Growing vulnerability disclosure and widespread software reuse increase security teams' need for reproducible evidence to diagnose vulnerabilities, validate patches, and build regression tests. Producing such evidence at scale requires automated end-to-end CVE reproduction. Existing methods...

Important: amazon-ssm-agent

Issue Overview: Handshakes which used Encrypted Client Hello could be de-anonymized by a passive network observer due to a disclosure of pre-shared key identities in the unencrypted client hello. CVE-2026-42505 go-git is an extensible git implementation library written in pure Go. Prior to 5.19.2...

Agent-Warden: EBPF-Based Kernel-Native Process-File Provenance Tracking for LLM Agents

LLM agents execute dynamically generated process and file operations that are often invisible to application-layer tracing. We present Agent-Warden, an extended Berkeley Packet Filter eBPF-based provenance monitor for tracking task and regular-file states across process creation, file access, and...

A Competing-Hazards Systematization of Loss of Control in Autonomous Agents

Leading AI developers have reported agents acting beyond their approved limits, which a United Nations panel described as an early warning of loss of human control. Yet incident reports and agent-safety evaluations describe these events differently, making it difficult to compare failures, trace...

Important: network-flow-monitor-agent

Issue Overview: libgit2 version 1.9.4 and below is vulnerable to a heap out-of-bounds read in setdata in src/libgit2/transports/smartpkt.c. The vulnerable code uses a fixed-size strncmp smartpkt.c:239 against the unvalidated capability buffer of a smart-protocol pkt-line. When the bytes following...

PT-2026-102686

Ollama versions 0.14.0 before 0.31.2 contain an incorrect authorization vulnerability in the experimental agent mode Bash tool approval mechanism that fails to properly parse shell syntax. Attackers who can influence model output through prompt injection can execute additional shell commands by...

Amazon Linux 2 : amazon-ssm-agent (ALAS-2026-3933)

The version of amazon-ssm-agent installed on the remote host is prior to 3.3.5226.0-1. It is, therefore, affected by multiple vulnerabilities as referenced in the ALAS2-2026-3933 advisory. Handshakes which used Encrypted Client Hello could be de-anonymized by a passive network observer due to a...

Xalgorix Autonomous AI Pentesting Agent 4.6.121

Most scanners detect. Xalgorix proves. An autonomous LLM agent works a full pentest methodology, then an independent verifier re-exploits every finding before it's reported - so you get proof, not a pile of maybes to triage. Self-hosted, private, and bring-your-own-LLM. Built in Go + TypeScript...

How we found 24 Android vulnerabilities using our open source AI security agent

With the rise of AI in the security space, our team created the GitHub Security Lab Taskflow Agent as a way for security researchers to easily automate, package, and share the AI prompts and workflows that they find effective for their work. In this blog post, I'll share how I created auditing...

0xtrails (>=0.0.0-20251106131028, <=0.18.6), 123pancli (1.0.0) +5780 more potentially affected by CVE-2026-101906 via axios (>=1.15.0, <=1.19.0)

axios NPM version =1.15.0, =0.0.0-20251106131028, =0.1.1, =0.0.0-develop-20260120180031, =8.13.1-beta.0, =6.13.1-beta-bump-deps.0, =0.0.0-canary-847463221a9a1bee28641d8c0ecfaca98ee142f6, =0.2.0, =0.0.1-alpha.9, =0.0.3-alpha.31, =0.0.1-alpha.29, =0.1.6-alpha.14, =0.0.1-alpha.20, =0.0.1-alpha.25 an...

CVE-2026-88808

A vulnerability has been identified within Rancher Manager where the Fleet agent wrote resources to downstream clusters using its own cluster-admin credentials instead of the ServiceAccount pinned to the deployment. It affects multi-tenancy environments where different tenants share the same...

EUVD-2026-88335

A vulnerability has been identified within Rancher Manager where the Fleet agent wrote resources to downstream clusters using its own cluster-admin credentials instead of the ServiceAccount pinned to the deployment. It affects multi-tenancy environments where different tenants share the same...