Docker Sandbox

When to use

• User selects "Docker Sandbox" as the execution environment in

  /replicate

  or

  /autoresearch

• Running untrusted code from a paper's repository
• Experiments that install packages or modify system state
• Any time the user asks to run something "safely" or "isolated"

How it works

1. Build or pull an appropriate base image for the research code
2. Mount the project directory into the container
3. Run experiment commands inside the container
4. Results write back to the mounted directory

Running commands in a container

docker run --rm -v "$(pwd)":/workspace -w /workspace python:3.11 bash -c "
  pip install -r requirements.txt &&
  python train.py
"

docker build -t feynman-experiment .
docker run --rm -v "$(pwd)/results":/workspace/results feynman-experiment

docker run --rm --gpus all -v "$(pwd)":/workspace -w /workspace pytorch/pytorch:latest bash -c "
  pip install -r requirements.txt &&
  python train.py
"

Choosing the base image

pytorch/pytorch:latest

tensorflow/tensorflow:latest-gpu

python:3.11

node:20

rocker/r-ver:4

julia:1.10

ubuntu:24.04

Persistent containers

/autoresearch

--rm

docker create --name <name> -v "$(pwd)":/workspace -w /workspace python:3.11 tail -f /dev/null
docker start <name>
docker exec <name> bash -c "pip install -r requirements.txt"
docker exec <name> bash -c "python train.py"

docker stop <name> && docker rm <name>

Notes

• The mounted workspace syncs results back to the host automatically
• Containers are network-enabled by default — add

  --network none

  for full isolation
• For GPU access, Docker must be configured with the NVIDIA Container Toolkit