External sandboxes for coding agents
- ID
- 47561
- Status
- summarized
- Published
- 18 Nov 2026, 6:52 AM
- Fetched
- 18 Nov 2026, 7:02 AM
- Provider
- Simon Willison
- Category
- developer-ai
- Original URL
- https://simonwillison.net/2026/Nov/17/external-sandboxes-for-coding-agents/
- Source URL
- https://simonwillison.net/atom/everything/
Summary
- Score
- 8.0
- Created
- 18 Nov 2026, 7:02 AM
- Tags
- Audience
- developersvibe_codersai_agent_users
What happened
Simon Willison introduced an external sandbox pattern where coding agents such as Claude Code, Codex, and Pi run outside a dedicated sandbox and send code-execution API calls into it. The project with Prime Radiant released three implementations: external-sandbox-v86 (Node.js plus v86/Alpine Linux in WebAssembly), external-sandbox-container (Python plus Alpine Linux in an Apple Containers instance), and external-sandbox-quicksand (Python using Microsoft's Quicksand library with QEMU and Alpine Linux). All expose the same localhost API, plus an optional HTTP/HTTPS allow-list proxy and credential replacement so agents can make authenticated calls without receiving the secret.
Why it matters
If you let Claude Code, Codex, or Pi execute code locally, this gives you a concrete isolation option with a host allow-list and secret-substitution proxy rather than exposing your machine and API keys directly. The practical decision is whether to point your agent at one of the three sandbox servers for untrusted or local-model workflows, especially when the agent needs network access but should not see credentials.
Discussion angle
Which agent workflows justify a network allow-list and credential proxy versus simply disabling network access, and what breaks when Claude Code or Codex must call an authenticated API from inside the sandbox?