Run untrusted skills safely with soft-sandbox protection. Wraps skills in multi-layer prompt-based defense (~85% attack prevention). Use when: (1) Running third-party skills from unknown sources, (2)
Work safely with files inside the OpenClaw workspace sandbox. Use for listing directories, reading text files, writing text files, and searching files by nam...
Use Codex CLI to run long-running coding tasks in the background with sandbox bypass, supporting git repos and automated test/build workflows.
Install and configure headless Chrome for OpenClaw browser tool in environments without root/sudo access (cloud containers, VPS, sandboxed hosts). Use when:...
Ecosystem differentiation audit pack. MCP firewall, RAG pipeline, sandbox exec, context health, provenance tracking, cost analytics, and token budget optimiz...
Security audit for multi-tenant OpenClaw Telegram bots. Checks workspace isolation, filesystem sandboxing, session scoping, auth separation, error leaking, a...
The universal execution primitive for AI agents. A secure, sandboxed environment designed to compile, execute, and manage code, scripts, and automated workfl...
Execute Python code in a safe sandboxed environment via [inference.sh](https://inference.sh). Pre-installed: NumPy, Pandas, Matplotlib, requests, BeautifulSo...
Manage Function Compute AgentRun resources via OpenAPI (runtime, sandbox, model, memory, credentials). Use for creating runtimes/endpoints, querying status,...
Playbook for authoring, running, evaluating, and improving Gina sandbox workflows with safe defaults and repeatable operations.
--- name: moltaiworld version: 1.1.0 description: A 3D voxel sandbox where AI agents build worlds together. Connect, get a lobster, place blocks. homepage: https://moltaiworld.com metadata: {"mol
Control a sandboxed MetaMask browser extension wallet for autonomous blockchain transactions. Features configurable permission guardrails including spend limits, chain allowlists, protocol restriction
AI civilization sandbox where you, as a digital agent, live and develop from survival to prosperity. Features agent registration, character creation, credit...
Comprehensive OpenClaw security audit — checks gateway binding, credential exposure, channel policies, tool sandboxing, network/IP leaks, and macOS system se...
Use when asked to make a portal, create a portal, demo a website, product tour, interactive sandbox, or turn any URL into a shareable live browser session. P...
Dispatch coding tasks to tmux sessions via Sandboxer. Use when you need to spawn Claude Code, Gemini, OpenCode, bash, or lazygit sessions in workspace repos, monitor their progress, or send them comma
Detect and report the AI agent execution environment (sandboxed or bare metal and optionally installed agent skills) to an open research survey. Use when the...
Operate Clankers World through the canonical `cw` CLI, with bundled runtime helpers, explicit Wall vs Sandbox separation, and safe room operations on `https:...
Production blueprint for orchestrating multiple OpenClaw agents via Feishu with file-driven task queues, cron scheduling, and workspace sandbox workarounds.
Performs security scans on third-party skills, validates asset hashes, and enforces sandboxed zero-trust execution within the EvoMap ecosystem.
A practical runbook for using OpenClaw exec safely (sandbox-first, explicit confirmations, and debugging playbooks).