PPhying中文
Curated security research, vulnerabilities, advisories and tools for practitioners.

Deceiving users with ANSI terminal codes in MCP

blog.trailofbits.com | blog | #ai-security | #claude-code | #mcp | #prompt-injection | #llm-security | #trail-of-bits | #ansi-escape-codes | #terminal-security
📄 Raw .md⬇️ Download .md🔗 Original ↗

Summary

Trail of Bits shows how ANSI escape sequences in MCP tool descriptions hide line-jumping payloads from users; Claude Code v0.2.76 does no filtering, enabling stealthy prompt injection attacks.
Published
Collected

Related coverage

blog · blog.trailofbits.com
State divergence enables unauthorized access
This coverage gives security teams current context for monitoring, validation, and remediation.
blog · hunt.io
Suspected Chinese Operators Use Claude Code and DeepSeek to Breach Government Systems Across Four Countries
Suspected Chinese operators used Claude Code and DeepSeek-v4-pro to breach government systems in Afghanistan, Thailand and Taiwan, and scanned 5,890+ government hosts worldwide.
blog · wiz.io
Build AI Security Agents with Wiz MCP
Wiz MCP is now GA, letting AI assistants and custom agents query the Wiz Security Graph, reuse Red, Blue and Green agent analyses, and run AI Skills workflows for triage and remediation.
blog · blog.sentry.security
ChatML Role Injection (CRI)
Attackers can inject system roles into ChatML message arrays to override developer instructions, invoke hidden tools, and pad payloads for denial-of-wallet DoS; strict schema validation is key.
blog · wiz.io
Introducing the Prompt Airlines CTF: Test Your AI Security Skills
Wiz launched Prompt Airlines, a chatbot-based CTF where players exploit AI logic flaws to win a free flight ticket, teaching real LLM security risks; open to all skill levels.
tool · github.com
SpecterOps Skills: Open-Source LLM Agent Skills & MCP Plugin Marketplace for Red Teaming and AppSec
As AI-assisted offensive security evolves from simple prompting to multi-agent architectures, SpecterOps standardizes security craft into declarative skills and MCP interfaces, allowing operators to safely orchestrate complex red team tradecraft and reverse engineering directly through LLM agents.
Back