First reported · updated · 15 reports redhat.com
Lead dispatch
First reported · updated · 3 reports embracethered.com
AWS Kiro: Arbitrary Code Execution via Indirect Prompt Injection
Researchers found a vulnerability (CVE-2026-10591) in AWS Kiro, an agentic IDE, where hidden instructions planted in a web page or source file that Kiro processes can trigger indirect prompt injection to rewrite Kiro's own MCP server configuration (~/.kiro/settings/mcp.json) or allowlist arbitrary Bash commands in .vscode/settings.json, achieving arbitrary code execution on the developer's machine with no approval prompt. The human-in-the-loop approval boundary is bypassed because Kiro can write to these config files without user consent, and AWS has issued a fix and CVE.indirect-prompt-injection · prompt-injection · remote-code-execution · tool-abuse · config-poisoning
ai-agents · mcp · llm · agentic-ide
Severity
0.85
The wire · latest
First reported github.com
GitHub - barvhaim/HoneyMCP: A Deception Security Layer for MCP Servers. It injects "ghost tools" (fake security-sensitive tools) that act as honeypots.
HoneyMCP is an open-source deception security layer for MCP (Model Context Protocol) servers that injects fake security-sensitive "ghost tools" acting as honeypots. When an agent or attacker invokes one of these decoy tools, the activity is flagged as suspicious and surfaced in a SOC-style dashboard, helping defenders detect malicious or compromised AI agents interacting with MCP servers. Details →First reported · updated · 2 reports adversa.ai
DeepJack Cursor deeplink vulnerability: 1-click MCP server RCE
Adversa AI disclosed a vulnerability in the Cursor AI coding IDE (dubbed "DeepJack") in which a crafted cursor:// deeplink can install an attacker-controlled MCP server that runs arbitrary, unsandboxed commands under the victim's account after one click and one confirmation. The install dialog renders the server command in a single-line field, pushing a malicious tail off-screen, and a double-URL-encoded variant disguises the mcp/install URI as a routine pr-review link. Cursor closed the reports as duplicates, but build 3.9.8 reportedly remains vulnerable. Details →First reported · updated · 2 reports atlan.com
Multi-Agent AI Security: 5 Compositional Risks and Fixes [2026]
An analysis piece on multi-agent AI security surveys compositional risks in agentic deployments — control-plane and orchestration-layer compromise, non-human identity gaps, credential persistence and scope creep, MCP server exposure, and static-permission failures — and proposes fixes like just-in-time least privilege and verification gates. The related arXiv paper systematically characterizes 193 MAS threat items across nine categories and evaluates 16 AI security frameworks, finding none achieves majority coverage of any single category and that Non-Determinism and Data Leakage are the most under-addressed. Details →First reported github.com
Incorrect authorization in the aggregation pipeline tool in Amazon AWS Labs DocumentDB MCP Server · Advisory · awslabs/mcp
A vulnerability (CVE-2026-18954, GHSA-j694-4m5j-w8hc) in Amazon AWS Labs DocumentDB MCP Server lets an authenticated MCP client bypass read-only mode enforcement by submitting aggregation pipelines with write-capable stages ($out, $merge) through the aggregate tool, which fails to validate pipeline stages. The issue is fixed in version 1.0.12; a workaround is to configure the server with a read-only database user. Details →First reported datadoghq.com
Before the first prompt: Code execution paths in trusted coding-agent projects | Datadog Security Labs
Datadog Security Labs demonstrates that trusting a repository in coding agents such as OpenAI Codex and Claude Code can trigger repository-controlled code execution before the first prompt, without a model response or shell-command approval. In Codex, project-scoped MCP configurations start an attacker-controlled process, and in Claude Code a project-controlled PATH causes Claude's automatic Git probes to run a tracked repository wrapper — bypassing hook-approval controls. Details →First reported · updated · 7 reports paloaltonetworks.com
Chinese-Speaking Threat Actor Harnesses AI Models for Autonomous Cyberattacks
Unit 42 reports a Chinese-speaking threat actor weaponized AI models—including DeepSeek and the open-source Hermes AI agent—to autonomously scan for, exploit, and compromise vulnerable servers, including an attack tied to the Thai Ministry of Finance. The actor drove the AI agent unattended via MCP tooling (FofaMap MCP server, Nuclei) to chain reconnaissance and exploitation of multiple CVEs with minimal human intervention. Details →First reported trussed.ai
MCP Security Vulnerabilities and Enterprise AI Agent
Trussed AI's write-up analyzes recurring structural security weaknesses in enterprise Model Context Protocol (MCP) deployments, including shared-credential identity, decentralized authorization with no central policy engine, dynamic tool discovery expanding prompt-injection surface, and weak auditability. It argues for external runtime governance—per-session scoped identity, a centralized policy engine intercepting tool calls, independent audit logging, and human approval for high-impact actions. Details →First reported google.com
Securing agentic AI: What's new in VPC Service Controls | Google Cloud Blog
Google Cloud announced new VPC Service Controls capabilities for securing agentic AI workloads, including treating AI agents as first-class IAM identities in perimeter ingress/egress rules, conditional access based on Model Context Protocol (MCP) attributes such as mcp.toolName and mcp.tool.isReadOnly, and native integration with the Gemini Enterprise Agent Platform that blocks public internet access. The features let administrators enforce least-privilege boundaries and revoke a compromised agent's access at the network perimeter. Details →First reported gitlab.com
LiteLLM: MCP Authentication Bypass via OAuth2 Passthrough Fallback | GitLab Advisory Database (GLAD)
CVE-2026-59822 is an authentication bypass in LiteLLM's MCP Streamable HTTP endpoint, where the OAuth2 passthrough fallback path replaced failed LiteLLM key validation with an empty UserAPIKeyAuth() object. An unauthenticated attacker could supply an arbitrary/fabricated Bearer token to establish an authenticated MCP session, then list and call configured MCP tools and access connected services. The issue is fixed in LiteLLM v1.84.0. Details →First reported mitiga.io
Modern Malware — Spyware Skills, Hijacked Base URLs, and 1,230+ Leaking API Keys in AI Instruction Files
Mitiga Labs details malware hidden in AI agent instruction files — Cursor rules, Anthropic Skills, Claude Hooks, AGENTS.md/CLAUDE.md context files, MCP server configs, and .pyc droppers — that AI agents follow with near-zero validation. The research found prompt-exfiltration tradecraft caught in the wild, attacker-controlled ANTHROPIC_BASE_URL overrides routing Claude traffic through MITM proxies, permission-bypass defaults, and over 1,230 hardcoded API keys and JWT tokens across tens of services. Mitiga also released a free scanner, Skillgate, built during the investigation. Details →First reported tenetsecurity.ai
A public Sentry key is all it takes to hijack Claude Code, Cursor, and Codex
Researchers at Tenet Security describe "agentjacking," an attack in which a publicly exposed Sentry key lets an attacker inject fake error messages that AI coding agents such as Claude Code, Cursor, and Codex ingest via the Sentry MCP server. The crafted error content acts as an indirect prompt injection, hijacking the agent to execute attacker-directed actions; the team also published a mitigation tool, agent-jackstop, on GitHub. Details →First reported strobes.co
CVE-2026-23744 - CVE Details, Severity, and Analysis | Strobes VI
CVE-2026-23744 is a critical (CVSS 9.8) remote code execution vulnerability in MCPJam Inspector versions 1.4.2 and earlier, a local-first development platform for MCP servers. Because the tool binds to 0.0.0.0 and its /api/mcp/connect endpoint extracts command and args without security checks, an attacker can send a crafted, unauthenticated HTTP request to trigger arbitrary command execution with no user interaction; public PoCs and vendor patches are available. Details →First reported mitiga.io
MCP Token Theft in Claude Code: A Man-in-the-Middle Attack Chain
Mitiga Labs research details a man-in-the-middle attack chain against Claude Code in which a user-level post-install hook rewrites MCP server endpoints in ~/.claude.json to route MCP traffic through attacker-controlled infrastructure and steal OAuth tokens for connected SaaS (Jira, Confluence, GitHub, etc.). Because provider-side audit logs still show valid OAuth traffic from Anthropic's trusted egress range, the malicious activity blends in as legitimate user actions, and token rotation fails to break the chain while the hook keeps reseeding the config. Details →First reported arxiv.org
The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities
A systematization-of-knowledge paper by Mohammadreza Rashidi organizes 39 works (2023-2026) on execution security for AI coding agents into 17 categories covering sandbox isolation, capability/access control, policy enforcement, TOCTOU races, and MCP threats, and verifies four disclosed, patched CVEs affecting production agent harnesses. It surfaces five cross-cutting gaps, including denylist policy failure rates of 69-98% and benign out-of-scope agent actions occurring at rates up to 17.1% under realistic prompting. Details →First reported microsoft.com
AutoJack: How a single page can RCE the host running your AI agent | Microsoft Security Blog
Microsoft researchers disclosed AutoJack, an exploit chain in AutoGen Studio where untrusted web content rendered by a browsing agent can reach a local MCP WebSocket and spawn arbitrary processes on the host, crossing the localhost trust boundary to achieve RCE. The issue was reported to MSRC and hardened upstream in commit b047730; the vulnerable MCP WebSocket surface was never shipped in a PyPI release. Details →How the wire is made
Poll & cluster
Internet is crawled for AI security news and near-duplicate coverage is embedded and grouped into durable items.
Curate
AI Agent filters for agentic-AI relevance, classifies and tags each item, scores severity for threats, and writes the summary.
Every item here is one machine-curated intelligence object, not a headline.
Read the wire for free. There is a small charge to ask the index questions.
The wire, open
The complete curated feed, no key required.
- GET /feed.xml — RSS 2.0, every item
- GET /api/items — read-only
The vector desk
Query the index by meaning, not just keyword.
- GET /api/items?tags=&minSeverity=&itemType=
- GET /api/search?q= — keyword
- GET /api/semantic?q= — vector