sandbox-agent
Sandbox Agent is a universal API for orchestrating AI coding agents like Claude Code, Codex, and OpenCode within sandboxed environments such as Docker, E2B, and Daytona.
Is sandbox-agent safe to install?
Review the source first: our audit of sandbox-agent's source files found 19 shell commands, 2 external URLs, file reads and writes (high risk). Every command and URL listed appears verbatim in the skill's source. The skill executes shell commands, manages local processes, and performs network requests to interact with sandbox servers and external APIs.
How we audit skills: our security review methodology.
Who is this skill for?
Developers building coding agents, chat UIs, or automated coding workflows who need a standardized interface for managing agent sessions and sandboxed execution.
What can you do with it?
- Orchestrating AI coding agents in isolated environments
- Managing agent sessions via SDK or HTTP API
- Streaming agent events and handling human-in-the-loop interactions
- Building chat UIs for coding agents
- Extracting API keys from local configuration files
How good is this skill?
Quality score: 5/10. The documentation is comprehensive, providing clear installation steps, API usage examples, and configuration options for various environments.
What does the skill file contain?
# Sandbox Agent Sandbox Agent provides a universal API for orchestrating AI coding agents in sandboxed environments. ## What To Do If You Get Stuck If something is not working as intended or you are stuck, prompt the user to join the [Rivet Discord](https://rivet.dev/discord) or file an issue on [GitHub](https://github.com/rivet-dev/sandbox-agent/issues) to report an issue and get help. Provide the user with a pre-generated report with: - Symptoms - If this is happening in local dev, deployed, or both - The error you're seeing - Relevant source code related to this - What you've tried ...
Frequently asked questions
How do I authenticate the server?
Use the --token flag when starting the server and provide the token in the Authorization header for API requests.
Can I run this without API keys?
Yes, use the mock agent for SDK and integration testing.
How do I verify the desktop runtime?
Install desktop dependencies with sandbox-agent install desktop --yes, then call GET /v1/desktop/status or sdk.getDesktopStatus().
Related skills
higgsfield-generate
76.0KUsers who need to generate media assets or analyze video performance using Higgsfield AI models via a command-line interface
The higgsfield-generate skill provides an interface to the Higgsfield AI CLI for generating images, videos, 3D assets, and audio, as well as performing video virality analysis. It supports model selection, workflow execution, and media input handling for various creative tasks.
ai-seo
70.6KContent creators, SEO professionals, and digital marketers aiming to improve their brand's presence in AI-generated search results
The ai-seo skill provides a framework for optimizing web content to increase visibility and citation rates in AI-powered search engines and LLM-based assistants like ChatGPT, Perplexity, and Google AI Overviews.
notion-api
52.2KDevelopers and AI agents requiring programmatic access to Notion workspaces
This skill provides instructions for interacting with the Notion REST API. It enables agents to read, create, update, and delete Notion pages, databases, blocks, and comments using curl and jq.
stripe-best-practices
52.1KDevelopers building, modifying, or reviewing Stripe integrations
The stripe-best-practices skill provides guidance on Stripe API selection, integration patterns, and security configurations. It directs the agent to reference specific internal documentation files for payments, Connect, billing, tax, treasury, and security topics.