- Exec Sandbox @inhuman/mcp-execSingle exec tool running caller Python in a network-isolated locked-down sandbox.11% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.1 GitHub stars.
- Runeward @runewardd/runewardGoverned execution cells for AI agents: Docker/K8s sandboxes with policy, audit, and approvals.3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.1 GitHub stars.
- Citadel @sustinbebustin/citadel-mcpA growing Citadel of stack-specific docs, served as MCP tools for AI coding agents.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
- Ag Bash @sairam0424/ag-bashSandboxed AI-native bash with 70 agentic tools: run_bash, code edit/diff, WASM Python/JS.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
- Rotifer Protocol MCP @rotifer-protocol/mcp-serverSearch, compare, and install self-evolving AI Agent Genes ranked by Arena fitness7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
- Python Exec Sandbox @lu-zhengda/mcp-python-exec-sandboxSandboxed Python execution for AI agents. PEP 723 inline deps, multi-version Python, zero pollution.8% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
- Declaw AI MCP @declaw-ai/mcp-serverSecure Firecracker microVM sandboxes for AI agents: network policy, PII & injection guardrails.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.1 GitHub stars.
- Reverse Engineering @daedalus/mcp-reverse-engineeringA sandboxed MCP tool for reverse engineering with multiple tool integrations7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.1 GitHub stars.
- Agent Workspace MCP @hrrodan/agent-workspace-mcpA sandboxed, agentic workspace providing secure filesystem, bash, and uv-powered Python execution.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.1 GitHub stars.
- fcvm @carlosbravoa/fcvmDisposable Firecracker microVM sandboxes for agents: run code, snapshot, fork. Needs Linux with KVM.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Spawner @flosk6/spawnerPreview environments on your self-hosted Spawner server: a copy of your app for each git branch3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- agend-sh @agend-sh/agend-shPersistent Linux environments for AI agents, with interactive terminals and HTTPS previews.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from Docker.
- macbox @opslane/macboxA real Mac for your agent: build, test, run, and tap through iOS apps on the simulator.Not assessed yet.Runs on your machine, installed from npm.
- Legwork @kumarganduri/legworkLets your AI find open-source tools on GitHub and install them as sandboxed MCP tools.Not assessed yet.Runs on your machine, installed from PyPI.
- Codex Subagent @parisbs/codex-subagent-mcpDelegate coding tasks to the OpenAI Codex CLI, installed separately, with explicit model and effort.Not assessed yet.Runs on your machine, installed from npm.
- chargehand @egoushka/chargehandRuns coding-agent workers on a codebase question; every claim in the answer carries evidence.4% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from NuGet.
- HackerEarth MCP @pratham-jain33/hackerearth-mcpGive any AI assistant a code run button via HackerEarth. 24 languages, zero setup.10% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- B2IGE Verify @b2ige2-a11y/b2ige-verifyIndependent deterministic verification for AI-written software. Local, evidence-backed checks.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.
- Capsulate @capsulatedev/capsulateReal Linux and Windows desktop VMs for AI agents: exec, files, computer-use and snapshots via MCP.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Verdict @dgotlieb/verdict-mcpSandboxed test runs for coding agents: compact typed verdicts with failure fingerprints and history3% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from PyPI.
- Soma @medianemi-beep/somaExecution-verified code generation and verification with signed, offline-checkable certificates.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Sandbox as a Service @sandbox-as-a-service/mcpGive an agent a real Linux VM: run commands, move files, expose a preview URL, destroy it.7% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from MCPB.
- Tenki Sandbox MCP @luxorlabs/tenki-mcpDisposable microVM sandboxes for AI agents: run code, read/write files, git, preview URLs.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
- Grok Build @nuruvala/grok-build-mcp-serverRun the Grok Build CLI as MCP tools: agent runs with live progress, code review, web research.5% chance this is malicious. Assessed by AI, which can make mistakes.Runs on your machine, installed from npm.
