CodeIntegrity blog

Research, exploit writeups, and practical notes on agent runtime risk.

Core controls

Platform overview
Data Flow Control
Tool Call Control
Policy Enforcement

Agent risks

All solutions
Prompt Injection
Data Leakage
Intent Drift
Memory Poisoning
Excessive Agency

Resources

All resources
Blog
Research

Company

About
Careers
Press Kit
Trust Center
Contact

Get Started

Newsletter

I agree to receive communications from CodeIntegrity.

© 2026 CodeIntegrity, Inc. San Francisco, California.

All Research Blog

prompt-injection

98% Accurate and Still Broken

StevenJan 3, 2026

The Hidden Risk in Notion 3.0 AI Agents: Web Search Tool Abuse for Data Exfiltration

AbiSep 19, 2025

Taint Analysis for Agent Tool Calls

StevenSep 1, 2025

Shopify Exploit: Manipulating Shoppers

AbiJul 14, 2025

Neon Exploit: Malicious SQL Injection

Azure Exploit: Leaking KeyVault Secrets

AbiJun 27, 2025

Heroku Exploit: App Ownership Takeover

AbiJun 15, 2025

Linear Exploit: Bypassing Team ACLs

AbiJun 14, 2025