george0papasotiriou/cve-2026-1337-ai-coding-assistant-prompt-injection-to-sandbox-escapeA prompt injection in a code‑review bot that executes AI‑generated fixes in a sandbox. The sandbox uses a blacklist to prevent dangerous commands, but a polyglot payload bypasses the filter and achieves remote code execution.
A prompt injection vulnerability in an AI-powered code review bot allows an attacker to inject arbitrary shell commands by manipulating the bot’s “fix” suggestion. The sanitizer uses a simple blacklist that is bypassed using encoding tricks.
##Fix: section. A blacklist attempts to block dangerous commands but fails against encoded payloads (base64, variable indirection).python ai_code_review_bot.py