AI Weekly Malaysia

Back to items Summaries

Claude Code and Gemini CLI Flaws Let a GitHub Issue Reach CI Workflow Secrets

ID
11882
Status
summarized
Published
07 Aug 2026, 4:18 PM
Fetched
07 Aug 2026, 5:39 PM
Provider
The Hacker News
Category
security
Original URL
https://thehackernews.com/2026/08/claude-code-and-gemini-cli-flaws-let.html
Source URL
https://feeds.feedburner.com/TheHackersNews

Summary

Score
8.5
Created
07 Aug 2026, 5:40 PM
Tags
Audience
developersvibe_codersai_agent_users

What happened

Novee Security demonstrated at Black Hat USA that a GitHub issue from an unprivileged account could execute code on CI runners behind Anthropic's, Google's, and OpenAI's own coding-agent repos. Gemini CLI's CVE-2026-12537 (CVSS 10.0) allows OS command injection via a crafted .gemini/.env file before the sandbox starts, fixed in Gemini CLI 0.39.1 and run-gemini-cli 0.1.22. Claude Code's CVE-2026-54316 used Hugging Face's public download counter as an API key exfiltration channel, fixed in 2.1.163 (all versions from 0.2.54 affected); OpenAI's Codex got no CVE, with OpenAI stating its sandbox behaved as documented.

Why it matters

If you run Claude Code or Gemini CLI in CI workflows that outside users can trigger via issues or PRs, update immediately to Claude Code 2.1.163, Gemini CLI 0.39.1, and run-gemini-cli 0.1.22, then audit which workflows accept untrusted input. The Gemini host-execution bug didn't even require prompt manipulation — a crafted .env file was enough. The deeper lesson: the harness code around the model (validators, allowlists, sandbox launchers) is where these failures cluster, not the model itself — Gemini CLI's tool allowlist was only checked at registration and ignored at runtime under --yolo.

Discussion angle

The harness is the attack surface: how to audit the code between your AI agent and the real world — validators that strip before checking, allowlists enforced only at registration, and sandbox boundaries that a .env file can cross before the sandbox even starts.

Top