Coding Assistants

Coding assistants are AI tools that generate, complete, or refactor source code based on natural language prompts or code context, with popular examples including GitHub Copilot and similar systems. For enterprise governance, these tools introduce risks around code quality, security vulnerabilities, license compliance, and intellectual property protection that require careful evaluation and oversight. Organizations must establish policies for which coding assistants are approved, how generated code is reviewed, and what guardrails prevent sensitive code from being transmitted to external AI systems.

1 item

ResearchGlobal2026-08-04

Poisoned AI Config Files Turn Sanctioned Coding Agents Into Exfiltration Tools

Security firm Mitiga has documented an attack technique called PromptLogger in which threat actors embed malicious instructions inside AI agent configuration files such as CLAUDE.md and .cursorrules to silently exfiltrate user prompts, credentials, and environment variables. Because the malicious behavior executes through the agent itself, it evades endpoint detection and response tools. The attack directly implicates enterprise AI governance programs that treat sanctioned coding assistants as low-risk approved software.

agentic AI supply chain attack data exfiltration coding assistants prompt injection AI governance credential theft