Research
Named attack vectors in the agent and MCP ecosystem: how each works, what it costs, and what stops it.
OCT 2026
Cross-Session Context Leakage
Sensitive data leaking between sessions, agents, and users through accumulated context windows, persistent memory, and shared server state.
SEP 2026
MCP Credential & Secrets Exposure
How plaintext credentials in MCP config files, environment variables, and error messages create a massive, largely unmonitored attack surface.
AUG 2026
MCP Observability & Audit Logging
Building comprehensive audit trails and real-time monitoring for all agent-to-tool interactions.
JUL 2026
Rug Pull Attacks in MCP
Trusted MCP servers that change behavior post-deployment to exfiltrate data or manipulate agent actions.
JUN 2026
Jailbroken AI
Techniques that bypass model safety layers through crafted MCP contexts, enabling restricted actions.
MAY 2026
Shadow AI
Unauthorized AI agents and MCP connections that operate outside IT visibility, bypassing security controls.
APR 2026
MCP Security Guardrails
Designing and deploying policy-driven constraints that keep AI agents operating within safe boundaries.
MAR 2026
Tool Poisoning Attacks
Malicious MCP servers that masquerade as legitimate tools to intercept sensitive data and corrupt agent behavior.
FEB 2026
Command Injection in MCP Tools
Exploiting unsanitized inputs in MCP tool calls to execute arbitrary commands on connected systems.
JAN 2026
MCP Endpoint & Inventory Sprawl
How unchecked growth of MCP servers creates hidden attack surfaces and operational blind spots across your AI infrastructure.