Docker Sandboxes isolate AI agents in microVMs with egress control and secret masking
mdelapenya · x · 2026-09-08
Testcontainers maintainer mdelapenya details how Docker Sandboxes (sbx) fit into agent architectures: sbx is a microVM with a specialized container runtime, a secrets manager, and network/filesystem policies to keep agents from harming the host.
A main agent like openclaw runs inside a DinD container so it can spawn additional containers on demand — e.g. for testcontainers-based tests (the author is a TC maintainer). The sandbox controls agent egress, allowing or blocking domains on the fly and substituting secrets so the agent never sees real credentials. Sandboxes are declared in compose-like YAML, with full docs available.
Related event: Docker Sandboxes Meets OpenClaw: Sandbox Boundaries Debated(5 posts)→
More from coding & agent
- xbrlkit: open-source XBRL layer over Arelle with built-in MCP server support — jfrench009 · 2026-09-09
- Herdr lead agent orchestrates sub-agents with worktree and PR skills, dev ditches the GUI — iannuttall · 2026-09-09
- Session acting weird? Open a new one: practitioners' fix for momentum prior and context rot — gerardsans · 2026-09-09
- Dev Adds a Complete Tech Tree With GPT-6 Astra in 2h21m — Dimillian · 2026-09-09
- Cheap models via OpenRouter fall apart in agentic harnesses: GLM and DeepSeek can't match Claude — scottyLogJobs · 2026-09-09
- Multi-agent coding's hardest problem: deciding who is allowed to change what — apghere · 2026-09-09