Topic

Prompt Injection: news, advisories and analysis

21 articlesUpdated 21 Oct 2026

CyberSecureToday covers Prompt Injection across 21 articles published between 9 Sep 2026 and 21 Oct 2026. The coverage includes 10 vulnerability advisories and 11 analysis or guidance pieces. 10 advisories are rated critical or high severity.

Stories on Prompt Injection most often intersect with Agentic AI, LLM Security and Model Context Protocol, with Technology, Cloud & SaaS the most affected sector. The most recent story, “Singapore CSA Releases National Cybersecurity Framework for Enterprise Frontier AI and Agentic Deployments”, was published 21 Oct 2026. Every article is listed below, newest first, with the affected products, severity and remediation guidance in each piece.

Latest coverage

InsightsCloud & SaaS
Singapore CSA Releases National Cybersecurity Framework for Enterprise Frontier AI and Agentic Deployments
CSA Singapore issues technical guidelines for securing frontier and autonomous agentic AI models, mandating rigorous defenses against indirect prompt injection and tool-calling escapes.
Mei-Lin Chen21 Oct 202614 min read
ThreatsCloud & SaaSHigh
AWS Bedrock Guardrails Vulnerability Enables Adversarial Safety Filter Evasion via Unicode Homoglyph Normalization (CVE-2026-80210)
AWS resolves a security bypass vulnerability in Amazon Bedrock Guardrails where adversarial Unicode homoglyph sequences evaded safety and PII masking filters.
CST Editorial Desk13 Oct 20268 min read
NewsCloud & SaaSCritical
Google Vertex AI Agent Builder Flaw Enables Sandbox Breakout and Token Exfiltration
Google patches a critical breakout flaw in Vertex AI Agent Builder (CVE-2026-76215) allowing recursive tool invocations to escape sandboxes and exfiltrate IAM tokens.
CST Editorial Desk12 Oct 20268 min read
ThreatsCloud & SaaSCritical
AWS Bedrock Agent Runtime Flaw Enabled Persistent Jailbreaks and Tool Hijacking
AWS patches CVE-2026-68115 in Amazon Bedrock Agent Runtime where unvalidated context memory serialization allowed cross-agent jailbreaks and tool invocation hijacking.
CST Editorial Desk11 Oct 20268 min read
ThreatsCloud & SaaSCritical
Anthropic Model Context Protocol Flaw Permits Indirect Prompt Injection Across Claude Enterprise Tool Callers
Anthropic and open-source MCP maintainers resolve CVE-2026-49110, an indirect prompt injection vulnerability allowing unauthorized tool execution in agentic pipelines.
Aisha Noor8 Oct 20268 min read
InsightsCloud & SaaS
Why Prompt Guards Cannot Stop Agentic RCE: The Case for MicroVM Isolation in Tool-Calling Pipelines
Recent CVEs in Ollama and AWS MCP reveal that semantic prompt filters cannot prevent tool injection. Engineering teams must isolate agent tools in ephemeral microVMs.
Aisha Noor6 Oct 20268 min read
ThreatsCloud & SaaSHigh
Ollama Patches Experimental Agent Mode Command Injection and Approval Bypass Flaw (CVE-2026-102697)
A security defect in Ollama experimental agent mode allowed command chaining operators to bypass confirmation gates and execute arbitrary shell commands.
Aisha Noor5 Oct 20268 min read
ResearchCloud & SaaS
OpenAI Disrupts Coordinated Adversarial Distillation Campaign Illicitly Harvesting Frontier AI Reasoning Traces
OpenAI dismantled a coordinated campaign that weaponized cross-model replay queries across 15,000 accounts to illicitly extract encrypted chain-of-thought traces.
Aisha Noor2 Oct 20269 min read
ThreatsCloud & SaaSHigh
Office PowerPoint MCP Server Path Traversal Enables Indirect Prompt Injection File Overwrite (CVE-2025-71427)
A path traversal vulnerability in the Office-PowerPoint Model Context Protocol server allows AI agents to be manipulated via prompt injection into overwriting host files.
Priya Raman2 Oct 20268 min read
ThreatsCloud & SaaSCritical
GitLab AI Gateway Prompt Execution Flaw Exposes Cloud Workspaces to Remote Code Execution (CVE-2026-90970)
A critical command injection flaw in GitLab AI Gateway allows remote attackers to break LLM agent sandboxes and execute code across enterprise cloud runners.
CST Editorial Desk1 Oct 202611 min read
ResearchCloud & SaaS
Self-Replicating AI Worms: Dissecting Autonomous Cross-Agent Prompt Injection Propagation
Researchers demonstrate Morris II style self-replicating adversarial prompt worms propagating autonomously across connected LLM agents and multi-agent workflows.
CST Editorial Desk30 Sep 202612 min read
ResearchCloud & SaaS
Cascading Failure in Multi-Agent AI Swarms: Indirect Prompt Injection Across Autonomous Tool-Calling Pipelines
Deep-dive security research demonstrating how indirect prompt injection exploits Model Context Protocol (MCP) to trigger cascading failures across multi-agent swarms.
Daniel Okafor28 Sep 202612 min read
InsightsCloud & SaaS
Model Context Protocol OAuth Passthrough Flaw Exposes Autonomous AI Tool Pipelines to Silent Impersonation
An architectural vulnerability in Model Context Protocol (MCP) OAuth token passthrough allows malicious tool servers to impersonate users and hijack enterprise APIs.
Priya Raman25 Sep 202618 min read
ThreatsCloud & SaaSHigh
BragJack Flaws Let Malicious Chrome and Edge Extensions Seize Agentic Browser Control
Security researchers demonstrate BragJack prompt forcing attacks against Chrome and Edge, hijacking agentic AI control planes to exfiltrate files and spy via camera.
Mei-Lin Chen23 Sep 202617 min read
ThreatsCloud & SaaSHigh
Amazon Bedrock AgentCore Flaw Allows Attackers to Bypass LLM Guardrails via Injected Tool-Use Payloads
AWS resolves high-severity vulnerability CVE-2026-18830 in Amazon Bedrock AgentCore allowing attackers to bypass LLM safety guardrails and trigger arbitrary tool calls.
Daniel Okafor19 Sep 202611 min read
ResearchCloud & SaaS
Langflow AI Pipeline Remote Code Execution Flaws Exploited in Enterprise LLM Deployments
CISA adds critical Langflow AI vulnerabilities CVE-2026-0770 and CVE-2026-55255 to KEV catalog as threat actors weaponize agentic pipeline execution flaws.
Aisha Noor18 Sep 202611 min read
ResearchCloud & SaaS
Agentic AI Security Breakdown: How Indirect Prompt Injections Hijack Autonomous Tool-Calling Agents
Analysis of emerging frontier AI threat vectors: how indirect prompt injections in untrusted documents hijack tool-calling autonomous LLM agents.
Priya Raman17 Sep 202614 min read
ThreatsCloud & SaaSCritical
Langflow AI Flaws Exploited to Steal Cloud API Keys and Deploy Miners
Vulnerability researchers observe mass automated scanning and exploitation of Langflow AI visual frameworks, exfiltrating OpenAI and AWS secrets while deploying botnets.
Priya Raman12 Sep 20266 min read
InsightsCloud & SaaS
The agentic AI security gap: 83% are deploying, 29% feel ready
Prompt injection is now OWASP’s top LLM risk with reported success rates of 50–84%. Organisations are shipping agents into production well ahead of their controls.
Priya Raman10 Sep 20268 min read
InsightsCloud & SaaS
88% Had an AI Agent Incident. 82% Believe Their Policy Covers It.
Gravitee says 88% of organisations had a confirmed or suspected AI agent incident, while 82% of executives trust their policy. Both figures, one population.
Priya Raman9 Sep 20266 min read
ResearchCloud & SaaS
DeepSeek Harness Vulnerability Allowed AI Coding Agents to Disable Their Sandbox
OX Security and VulnCheck disclosed CVE-2026-82533 in DeepSeek Harness, where prompt injection allowed sandboxed AI coding agents to disable their own security isolation.
Mei-Lin Chen9 Sep 20265 min read