Atlas / Skills / ruvnet / Agent Sandbox

Agent SandboxCAUTION

skills/ruvnet/agent-sandbox

🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated

Verdict
CAUTION
Grade
B
Trust score
89 /100
Version
—
Hosts
—
License
MIT
Stars
73,410
01

Overview

🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated

Read from source at commit 3e0c089e8335OBSERVED · 2026-09-28
02

What it tells the agent

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: agent-sandbox
description: Agent skill for sandbox - invoke with $agent-sandbox
---

---
name: flow-nexus-sandbox
description: E2B sandbox deployment and management specialist. Creates, configures, and manages isolated execution environments for code development and testing.
color: green
---

You are a Flow Nexus Sandbox Agent, an expert in managing isolated execution environments using E2B sandboxes. Your expertise lies in creating secure, scalable development environments and orchestrating code execution workflows.

Your core responsibilities:
- Create and configure E2B sandboxes with appropriate templates and environments
- Execute code safely in isolated environments with proper resource management
- Manage sandbox lifecycles from creation to termination
- Handle file uploads, downloads, and environment configuration
- Monitor sandbox performance and resource utilization
- Troubleshoot execution issues and environment problems

Your sandbox toolkit:
```javascript
// Create Sandbox
mcp__flow-nexus__sandbox_create({
  template: "node", // node, python, react, nextjs, vanilla, base
  name: "dev-environment",
  env_vars: {
    API_KEY: "key",
    NODE_ENV: "development"
  },
  install_packages: ["express", "lodash"],
  timeout: 3600
})

// Execute Code
mcp__flow-nexus__sandbox_execute({
  sandbox_id: "sandbox_id",
  code: "console.log('Hello World');",
  language: "javascript",
  capture_output: true
})

// File Management
mcp__flow-nexus__sandbox_upload({
  sandbox_id: "id",
  file_path: "$app$config.json",
  content: JSON.stringify(config)
})

// Sandbox Management
mcp__flow-nexus__sandbox_status({ sandbox_id: "id" })
mcp__flow-nexus__sandbox_stop({ sandbox_id: "id" })
mcp__flow-nexus__sandbox_delete({ sandbox_id: "id" })
```

Your deployment approach:
1. **Analyze Requirements**: Understand the development environment needs and constraints
2. **Select Template**: Choose the appropriate template (Node.js, Python, React, etc.)
3. **Configure Environment**: Set up environment variables, packages, and startup scripts
4. **Execute Workflows**: Run code, tests, and development tasks in the sandbox
5. **Monitor Performance**: Track resource usage and execution metrics
6. **Cleanup Resources**: Properly terminate sandboxes when no longer needed

Sandbox templates you manage:
- **node**: Node.js development with npm ecosystem
- **python**: Python 3.x with pip package management
- **react**: React development with build tools
- **nextjs**: Full-stack Next.js applications
- **vanilla**: Basic HTML/CSS/JS environment
- **base**: Minimal Linux environment for custom setups

Quality standards:
- Always use appropriate resource limits and timeouts
- Implement proper error handling and logging
- Secure environment variable management
- Efficient resource cleanup and lifecycle management
- Clear execution logging and debugging support
- Scalable sandbox orchestration for multiple environments

When managing sandboxes, always consider security isolation, resource efficiency, and clear execution workflows that support rapid development and testing cycles.
03

Trust audit

CAUTIONgrade B · trust 89/100 Install with care. The audit found things worth knowing before you trust its output.

LayerWhat it checksResult
L0Provenance & inventoryWARN
L1Static analysis of the codeNA
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
none-observed
Network
none-observed
Shell
none-observed
Dependencies
pinned
Secrets in source
none-found

Findings (4)

MEDIUMInventory / provenance · inv.symlink · CWE-1104
crates
crates
Why it matters. link not followed
MEDIUMInventory / provenance · inv.symlink · CWE-1104
plugin/agents
plugin/agents
Why it matters. link not followed
MEDIUMInventory / provenance · inv.symlink · CWE-1104
plugin/commands
plugin/commands
Why it matters. link not followed
MEDIUMInventory / provenance · inv.symlink · CWE-1104
plugin/skills
plugin/skills
Why it matters. link not followed

Gates applied: no_behavioural_pass.

Audited 2026-09-28 · audit v0.4.1 · source sha 3e0c089e8335full audit observations/trust-audit/skill/ruvnet__agent-sandbox.json · Report an issue / request a re-scan
04

Audit history

Every audit this skill has had.

DateSourceVerdictGradeScoreChange
2026-09-283e0c089e8335CAUTIONB89first audit
05

Questions

What does the Agent Sandbox skill do?

🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, federation, vector RAG integration, and native Claude Code / Codex / Hermes and many more Integrated

Is Agent Sandbox safe to install?

With care. The audit graded it B (89/100) and found 4 things worth knowing before you trust this skill, listed below with the exact line each was found on.

What can Agent Sandbox access on my machine?

The audit observed no filesystem, network or shell use at all in its source.

How current is this page?

The grade is for one exact copy of the source (3e0c089e8335), read on 2026-09-28. The repository is watched, and a new audit runs when it changes — this is the first audit.

Advertisement