PromptPwnd: Prompt Injection Vulnerabilities in GitHub Actions Using AI Agents
· Source: aikido.dev
In short
Aikido Security discovered a new class of vulnerabilities called PromptPwnd affecting GitHub Actions and GitLab CI/CD pipelines when integrated with AI agents like Gemini CLI, Claude Code, and OpenAI Codex. The vulnerability allows attackers to inject malicious prompts through untrusted user input, leading to secret exfiltration and workflow manipulation. At least 5 Fortune 500 companies are impacted, with Google's Gemini CLI repository being one of the confirmed cases that was patched within four days of disclosure.
Read Original Article — Read the full article on the publisher’s site.
In the issue of week 49, 2025: Agentic AI Orchestrates Supply Chains as Security and Governance Accelerate
Tags: security, automation, vulnerability, ci-cd, prompt-injection, supply-chain