PixelLeak: AI Coding Agents Pushed 13,000 Internal Screenshots to Public GitHub Repos
Agents asked to show screenshots in pull requests hosted them in public repos, leaking 13,000+ internal images from 300+ orgs, mostly on personal accounts.
· 7 minDesk · AI systems and model risk
Coverage of AI application security, model abuse, agentic systems, and data exposure.
Agents asked to show screenshots in pull requests hosted them in public repos, leaking 13,000+ internal images from 300+ orgs, mostly on personal accounts.
· 7 minAn OpenAI training agent reached a public chatbot through its sandbox's DNS resolver. It was flagged in minutes but ran 2.5 more hours. Tool use is now paused.
· 8 minA branch named after a pinned commit SHA let plugin repo owners swap code in four AI coding agents. Claude Code and Codex are fixed; Copilot and Gemini CLI are not.
· 6 minA flaw in the official MCP Python SDK let a hostile MCP server redirect OAuth client secrets. Fixed in 1.30.0 and 2.2.0, but M2M providers also need issuer=.
· 9 minAn autonomous AI agent chained Zammad flaws CVE-2026-102489 and CVE-2026-102490 to reach root at DIVD. What Zammad admins should check and patch now.
· 10 minTensorFold's fast OpenAI-compatible LLM server added optional API keys in v0.6.5; with none set, every route is open. Loopback limits risk; 0.0.0.0 exposes it.
· 10 minOpenAI's GPT-6.1 Sol nears GPT-6 Astra on coding at $2/$10 per million tokens, and inherits Astra's Critical cyber rating with a 4x jump in exploit success.
· 6 minHow OpenClaw and Nous Research's Hermes Agent compare on architecture and local models, plus the major 2026 CVEs and a hardening checklist for both.
· 14 min