sandbox-use

Run agent tasks inside isolated Docker sandboxes with per-space mounts.

1|2|Updated Apr 29, 2026
One-click install
npx skills add https://github.com/fuzzy-dynamics/strings --skill sandbox-use
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sandbox-use
Source: https://github.com/fuzzy-dynamics/strings/tree/main/skills/sandbox-use
Command: npx skills add https://github.com/fuzzy-dynamics/strings --skill sandbox-use

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires docker, jq, and includes scripts (resource) components.

What problem does it solve?

OpenScientist agents need isolated, reproducible execution environments to access tools and custom toolchains without polluting the host. This skill provides managed Docker sandboxes that can be concurrently active, bound to per-space mounts, with clear activation semantics and workdir transparency.

Core Features & Use Cases

  • Per-space sandbox isolation: run toolchains inside containers bound to the agent's workspace, ensuring file path parity with the host.
  • Dynamic activation: activate.sh --mount <path> or deep-run activations allow agents to attach or attach-on-demand to sandboxes.
  • Workdir-aware execution: exec.sh enforces binding constraints and preserves host-dir semantics, enabling predictable behavior for host tools and agent workflows.

Quick Start

Activate a sandbox and run a command inside it for your agent workflow.

Frequently Asked Questions about sandbox-use

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run agent tools in isolated Docker sandboxes?

Run agent tools in isolated Docker sandboxes by using dynamic activation scripts like activate.sh to mount per-space paths and exec.sh to enforce binding constraints for reproducible execution.

Why do I need sandbox isolation for OpenScientist agent workflows?

Sandbox isolation prevents agents from polluting the host environment by providing reproducible, secure Docker containers that ensure file path parity and predictable toolchain execution.

Can I bind custom toolchains to a specific workspace mount in Docker?

Yes, you can bind custom toolchains to specific workspace mounts using activate.sh --mount <path>, which ensures per-space isolation and workdir transparency for agent file operations.

Do I need Docker and jq installed to use sandbox orchestration scripts?

Yes, Docker and jq are required dependencies for sandbox orchestration, as Docker provides the container execution environment and jq parses live container state from docker inspect outputs.

How does workdir-aware execution keep host directory paths consistent?

Workdir-aware execution preserves host directory semantics inside the Docker sandbox by enforcing binding constraints through exec.sh, ensuring agents experience predictable file path parity.

What are the limitations of running concurrent multi-sandbox Docker deployments?

Concurrent multi-sandbox deployments require careful per-space mount management and dynamic activation to avoid binding conflicts, relying on docker inspect for live state tracking across containers.