Containment for AI agents - user isolation, sandboxed execution, network controls, backup/rollback. TLA+ verified.

Run multiple agents in parallel sandboxed VMs, with a single command, on your PC or in the cloud

Local-first, self-hosted AI agent runtime and MCP bridge with sandboxed sessions, memory, credentials, audit/replay, and a local Console.

Open-source sandboxed agent harness for teams. Giving every employee a secured personal agent.

Vigilante is a sandbox-first orchestration layer for coding agents. It isolates every task in a git worktree, enforces strict credential scoping, and gives you full audit logs — so your agents can't burn down production.

🛠️ The meta-harness for AI agents — scaffold your own focused, branded agent harness with its own npx CLI, MCP server, memory, learning loop, and witness-signed releases. Works with Claude Code, Codex, pi.dev, Hermes, OpenClaw, and RVM (hardware-isolated sandbox).

Self-hosted AI agent harness in a single Go binary — writes, sandbox-tests and repairs its own tools, and lets Claude Code, Codex and any MCP client build and share them.

Playwright for coding agents. Benchmark Claude Code, Codex, Gemini, and OpenCode on your own tasks - and test that your skills, MCP servers, and CLIs work when an agent uses them. Sandboxed YAML suites, activation checks, A/B experiments, CI gates.

A practical framework for AI-Assisted Research in Mathematics and Machine Learning

Safe local execution layer for AI agent tools. Build, validate, and publish MCP tools with a no-pass-no-run workflow — cross-platform desktop app powered by Spring AI.