Blocking at the Boundary: Auditing Long-Horizon Agents against Staged Prompt Injection

Long-horizon agents consume external content, invoke tools, and modify persistent state. Indirect prompt injection can exploit task-specific context, propagate across causally connected stages, and alter a consequential action while the workflow continues; we term this staged prompt injection. We...

Who Is Your Agent Serving? Provider-Side Indirect Prompt Injection in Proactive Agents

Proactive personal agents increasingly decide what to recommend, how to personalize advice, and what follow-up assistance to offer, creating a new user-decision attack surface for provider-side indirect prompt injection. We show that an external provider need not access private user context,...

anticipator

Anticipator Seguridad en tiempo de ejecución para sistemas de IA multiagente. Anticipator detecta inyección de instrucciones, filtración de credenciales, ataques de codificación, suplantación de homoglifos, path traversal y comportamiento anómalo de agentes en pipelines de LangGraph — antes de qu...

APEX: Active Protection at Execution Boundaries for LLM Agents

Indirect prompt injection IPI hides adversarial instructions in content that large language model LLM agents read at runtime. As agents compose heterogeneous capability units, including Tools, MCP servers, and Skills, the carriers of injection multiply, and defenses built to recognize attack...

Prompt-Injection-in-the-Wild

Prompt Injection in the Wild A tracker of publicly reported prompt-injection techniques from roughly the last two years, maintained by Rachel James cybershujin. Each technique is broken down into the three elements of a prompt injection: 1. Delivery method — how the injected instructions reach th...

bordair-detector

Bordair Detector A two-stage detector for prompt injection and jailbreak attempts in LLM inputs. A regex gate settles the easy majority of traffic without touching the model; anything ambiguous falls through to a quantised DeBERTa-v3 classifier running in ONNX. The same engine covers image,...

CVE-2026-90970

GitLab has remediated a vulnerability in the GitLab AI Gateway component affecting all versions of the AI Gateway from 18.1.6 before 19.2.4, 19.3 before 19.3.2, and 19.4 before 19.4.1 that, under certain conditions, could have allowed an authenticated user with Duo Agent Platform access to escape...

CVE-2026-90970 Improper Neutralization of Special Elements Used in a Template Engine in GitLab AI Gateway

GitLab has remediated a vulnerability in the GitLab AI Gateway component affecting all versions of the AI Gateway from 18.1.6 before 19.2.4, 19.3 before 19.3.2, and 19.4 before 19.4.1 that, under certain conditions, could have allowed an authenticated user with Duo Agent Platform access to escape...

EUVD-2026-91651

GitLab has remediated a vulnerability in the GitLab AI Gateway component affecting all versions of the AI Gateway from 18.1.6 before 19.2.4, 19.3 before 19.3.2, and 19.4 before 19.4.1 that, under certain conditions, could have allowed an authenticated user with Duo Agent Platform access to escape...

CVE-2026-90970 Improper Neutralization of Special Elements Used in a Template Engine in GitLab AI Gateway

GitLab has remediated a vulnerability in the GitLab AI Gateway component affecting all versions of the AI Gateway from 18.1.6 before 19.2.4, 19.3 before 19.3.2, and 19.4 before 19.4.1 that, under certain conditions, could have allowed an authenticated user with Duo Agent Platform access to escape...

CVE-2026-90970

GitLab has remediated a critical vulnerability in the GitLab AI Gateway component (CWE-1336) affecting versions 18.1.6 before 19.2.4 , 19.3 before 19.3.2 , and 19.4 before 19.4.1 . An authenticated user with Duo Agent Platform access could exploit a template engine weakness to escape the prompt t...

CVE-2026-90970 Improper Neutralization of Special Elements Used in a Template Engine in GitLab AI Gateway

GitLab has remediated a vulnerability in the GitLab AI Gateway component affecting all versions of the AI Gateway from 18.1.6 before 19.2.4, 19.3 before 19.3.2, and 19.4 before 19.4.1 that, under certain conditions, could have allowed an authenticated user with Duo Agent Platform access to escape...

EUVD-2025-211024

Office-PowerPoint-MCP-Server through 2.0.7 contains a path traversal vulnerability that allows MCP callers to write and read files outside the working directory by supplying absolute paths or ../ sequences. Attackers can steer an AI agent via prompt injection to abuse savepresentation,...

CVE-2026-90970: Improper Neutralization of Special Elements Used in a Template Engine

GitLab has remediated a vulnerability in the GitLab AI Gateway component affecting all versions of the AI Gateway from 18.1.6 before 19.2.4, 19.3 before 19.3.2, and 19.4 before 19.4.1 that, under certain conditions, could have allowed an authenticated user with Duo Agent Platform access to escape...

PT-2026-104355

Name of the Vulnerable Software and Affected Versions GitLab AI Gateway versions 18.1.6 through 19.2.3 GitLab AI Gateway versions 19.3 through 19.3.1 GitLab AI Gateway versions 19.4 through 19.4.0 Description An improper neutralization weakness in the GitLab AI Gateway allows an authenticated use...

Securing Computer-Use Agents against Branch Steering Attacks

Modern Computer Use Agents CUAs directly interact with graphical user interfaces and execute third-party web tools, exposing them to indirect prompt injection across every rendered page and tool response. While the Dual-LLM pattern is the primary system-level architecture offering formal security...

Containing the Autonomous Operator: A Defense-In-Depth Framework and Reference Architecture for Securing AI Agents on Kubernetes

Large language model LLM agents are moving from chat interfaces into infrastructure operations, where they read telemetry, call tools, generate and execute code, and change the state of production Kubernetes clusters. This collapses a boundary that conventional cloud-native security assumes: the...

CVE-2025-71427 Office-PowerPoint-MCP-Server through 2.0.7 Path Traversal via save_presentation and manage_image

Office-PowerPoint-MCP-Server through 2.0.7 contains a path traversal vulnerability that allows MCP callers to write and read files outside the working directory by supplying absolute paths or ../ sequences. Attackers can steer an AI agent via prompt injection to abuse savepresentation,...

CVE-2025-71427 Office-PowerPoint-MCP-Server through 2.0.7 Path Traversal via save_presentation and manage_image

Office-PowerPoint-MCP-Server through 2.0.7 contains a path traversal vulnerability that allows MCP callers to write and read files outside the working directory by supplying absolute paths or ../ sequences. Attackers can steer an AI agent via prompt injection to abuse savepresentation,...

CVE-2025-71427 Office-PowerPoint-MCP-Server through 2.0.7 Path Traversal via save_presentation and manage_image

Office-PowerPoint-MCP-Server through 2.0.7 contains a path traversal vulnerability that allows MCP callers to write and read files outside the working directory by supplying absolute paths or ../ sequences. Attackers can steer an AI agent via prompt injection to abuse savepresentation,...