# Running Pulumi Neo in a Docker Sandbox: workshop recap

DevFeed: [Running Pulumi Neo in a Docker Sandbox: workshop recap](<https://devfeed.tech/articles/running-pulumi-neo-in-a-docker-sandbox-workshop-recap-92883.md>)

Original publisher: [Read original article](<https://www.pulumi.com/blog/running-pulumi-neo-in-a-docker-sandbox/>)

Author: Mike Coleman

Published: 2026-10-21T00:00:00Z

Content type: article

Language: en

Sources: [Pulumi](<https://devfeed.tech/sources/pulumi.md>)

Topics: [pulumi-neo](<https://devfeed.tech/topics/pulumi-neo.md>), [Docker Sandboxes](<https://devfeed.tech/topics/docker-sandboxes.md>), [pulumi](<https://devfeed.tech/topics/pulumi.md>), [Docker](<https://devfeed.tech/topics/docker.md>), [AI Infrastructure](<https://devfeed.tech/topics/ai-infrastructure.md>), [personal access token](<https://devfeed.tech/topics/personal-access-token.md>), [World models](<https://devfeed.tech/topics/world-models.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [api](<https://devfeed.tech/tags/api.md>), [coding-agent](<https://devfeed.tech/tags/coding-agent.md>), [community](<https://devfeed.tech/tags/community.md>), [data](<https://devfeed.tech/tags/data.md>), [docker](<https://devfeed.tech/tags/docker.md>), [docker-sandbox](<https://devfeed.tech/tags/docker-sandbox.md>), [esc](<https://devfeed.tech/tags/esc.md>), [infrastructure](<https://devfeed.tech/tags/infrastructure.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [neo](<https://devfeed.tech/tags/neo.md>), [policy](<https://devfeed.tech/tags/policy.md>), [production](<https://devfeed.tech/tags/production.md>), [prompt](<https://devfeed.tech/tags/prompt.md>), [proxy](<https://devfeed.tech/tags/proxy.md>), [pulumi](<https://devfeed.tech/tags/pulumi.md>), [pulumi-neo](<https://devfeed.tech/tags/pulumi-neo.md>), [recap](<https://devfeed.tech/tags/recap.md>), [sandbox](<https://devfeed.tech/tags/sandbox.md>), [security](<https://devfeed.tech/tags/security.md>), [token](<https://devfeed.tech/tags/token.md>), [workshop](<https://devfeed.tech/tags/workshop.md>)

## AI overview

The workshop describes running Pulumi Neo in a Docker Sandbox with controls outside the agent's prompt. Neo runs in an isolated microVM with access limited to mounted files and permitted network hosts; a proxy injects the Pulumi token so Neo does not hold the real credential. Pulumi governs permissions and approvals: manual asks before each tool call, balanced before `pulumi up`, and auto never asks. Read-only blocks Pulumi Cloud writes but does not prevent access to ESC environments or their cloud accounts; role permissions must restrict environment access. Preventative Pulumi Policies can block violations, `protect: true` can stop deletion of protected resources, and ESC can issue short-lived AWS credentials, which limit leak damage but do not prevent it. Docker's network proxy can deny unapproved hosts, though allowed hosts still receive what Neo sends. The excerpt recommends starting with a locked-down network preset and balanced approvals near production. Centralized Docker sandbox controls are paid; local sandboxes and the `sbx` CLI are free. The final teardown demonstration was not completed.

## Source excerpt

The more an agent can do on its own, the more infrastructure work you can hand it, as long as its limits hold when it gets something wrong. In April, they didn't hold for PocketOS. A coding agent on a routine staging task deleted the company's production volume in 9 seconds. According to founder Jer Crane's account, it used an API token it found in an unrelated file. Its explanation afterward: "I guessed that deleting a staging volume via the API would be scoped to staging only. I didn't verify." Railway recovered the data two days later. The agent had been told not to. "System prompts are advisory, not enforcing," Crane wrote afterward. If the prompt can't stop an agent, the limits have to sit where the agent can't negotiate with them. That was the premise of Neo in a Docker Sandbox, the workshop Adam Gordon Bell and I ran with Mike Coleman from Docker. So we gave Pulumi Neo two layers. The first is a Docker Sandbox. Neo runs in its own microVM and reaches only the files and hosts we allow. The Pulumi token gets injected by a proxy on the way out, so Neo never holds the real credential. The second is Pulumi itself, which decides what Neo may change and when it has to ask. With both in place, you can let Neo work without approving every step. For why YOLO mode belongs in a sandbox at all, see my August post. Mike Coleman is a Principal Solutions Architect at Docker. His running joke is that agents "want to be helpful". It happens in his own demos: Mike Coleman I talk to people about agents a lot. They just really want to please you. I don't know how many times I've done a demo [of] governance for MCP servers, and I'll be like, "Can you write to my Notion page?" And there will be a policy applied to the sandbox that says it can't. But I have a connector enabled in Claude somewhere. And it's like, "Well, not with that MCP server, but I can use this thing over here to do it." Mike's answer is to close those doors before the agent finds them: Mike Coleman The more you c