17,600 Actions: Agent Security Is a Systems Problem

Summary

Security incidents involving AI agents, such as the OpenAI and Hugging Face event where attackers executed 17,600 actions, demonstrate that system security cannot rely entirely on manual human review. Managing agent velocity requires dedicated system-level controls to properly constrain, observe, and govern their behaviors during execution. To secure autonomous development workflows, Docker Sandboxes allow AI agents to operate within GitHub Actions, run Testcontainers tests, adjust code, and automatically generate draft pull requests inside isolated environments. In addition, escalating software supply-chain threats and the increased volume of AI-generated code drive the adoption of Docker Hardened Images and Docker Extended Lifecycle Support. These updates incorporate software built from source into container images, extend security patch coverage past end-of-life milestones, and move organizational policy enforcement directly onto developer machines.

Context

The OpenAI and Hugging Face incident involved 17,600 attacker actions, demonstrating that AI agent security cannot depend on human review. In parallel, software supply-chain attacks are escalating while AI generates more production code, alongside software components reaching end of life.

Approach / What changed

Implement controls to constrain, observe, and govern AI agents; run isolated agents with Docker Sandboxes in GitHub Actions to execute Testcontainers tests, remediate code, and create draft pull requests; and use Docker Hardened Images and Docker ELS to build images from source, extend end-of-life security coverage, and enforce policies on developer machines.

Takeaways

• An incident involving OpenAI and Hugging Face recorded 17,600 attacker actions, proving that AI agent security cannot rely on human review.
• Docker Sandboxes provide isolated execution environments in GitHub Actions for AI agents to run Testcontainers tests, fix code, and open draft pull requests.
• Docker platform updates add software built from source, provide security coverage past software end of life, and enforce policies on developer machines.

Related reading

• Docker · Aug 18, 2026

  Coding Agent Horror Stories: The Command You Already Approved

  The OpenAI and Hugging Face security incident involved 17,600 attacker actions, demonstrating that AI agent security cannot rely exclusively on manual human review. Modern development workflows require systems controls to constrain, observe, and govern autonomous agents operating at high speeds. To address related isolation and automation requirements, developers can run AI agents inside GitHub Actions using Docker Sandboxes to execute Testcontainers tests, fix code, and generate draft pull requests. In addition, Docker Extended Lifecycle Support provides continued security coverage past software end of life while moving policy enforcement directly to developer workstations. Finally, Docker Hub has transitioned Docker Verified Publisher applications to a self-serve model, enabling publishers to distribute verified content directly to developers.

  Jennifer Kohl

• Docker · Jul 20, 2026

  Coding Agent Horror Stories: The Agent That Deleted Production