Catalogue
Submit a toolHarnesses, frameworks, tools, apps, and platforms for agent builders, each scored on real GitHub credibility and each with a credibility-gated forum.
47 tools · showing 25–47
ToolSandcastle
Coding
mattpocock
TypeScript library for running AI coding agents in isolated Docker, Podman, or Vercel sandboxes and merging commits back
dockersandboxgitorchestration
ToolHarnessRouter
DeploymentInterface
HarnessRouter
HarnessRouter Community Edition: the self-hosted, Apache-2.0 edition of the unified interface for agent harnesses. Run Codex, Claude Code, H
gatewayuhpsandboxcoding-agent
ToolOpen Cowork
CodingInterface
OpenCoworkAI
Open-source AI agent desktop app for Windows & macOS. One-click install Claude Code, MCP tools, and Skills — with sandbox isolation, multi-m
desktopsandboxmulti-providercoding-agent
Smolagents
Coding
huggingface
Hugging Face library for agents that write their actions as Python code, with sandboxed execution
huggingfacecoding-agentmultimodalsandbox
ToolPi Go
CodingInterface
dimetron
Go implementation of AI coding agent
coding-agentclilspmulti-agent
Sandbox
SecurityInterface
agent-infra
All-in-One Sandbox for AI Agents that combines Browser, Shell, File, MCP and VSCode Server in a single Docker container.
sandboxcode-executionbrowser-automationdocker
HarnessSandBase Harness
SecurityMonitoring
sandbaseai
Open-source CMA-compatible agent runtime for any model, with MCP tools, sandboxed sessions, audit, replay, and a local console. Includes a n
sandboxlocal-firstgovernancesession-replay
Cloudflare OS
InterfaceSecurityCoding
cloudflare
Agent workspace built on Cloudflare Workers for creating documents, building apps, and running agents with your company’s context and system
productivitysandboxguardrailsapp-generationhuman-in-the-loop
ToolSpacedock
CodingQA
spacedock-dev
Pi ships four tools ... read, write, edit, bash ... and calls it done. No MCP, no sub-agents, no permission dialogs, no
human-in-the-loopmulti-agentorchestrationgovernance
ToolAos Ce
DeploymentSecurity
unicity-aos
AOS Community Edition: the open agent operating system.
clisandboxdaemon
ToolShepherd
Coding
shepherd-agents
Records agent runs as reversible, inspectable execution traces whose outputs are reviewed before they are applied or discarded
sandboxagent-builderreversible-executionmeta-agent
PlatformContainer use
CodingSecurity
dagger
MCP server giving each coding agent an isolated container and git branch so multiple agents work in parallel without clobbering each other.
daggersandboxdockergit-branching
HappyCapy
InterfaceCoding
Browser-based agent workspace running skills on frontier models under one subscription, with async tasks landing in an inbox
sandboxcoding-agentskillbrowser-automation
ToolOpen Computer Use
Interface
e2b-dev
AI computer use powered by open source LLMs and E2B Desktop Sandbox
computer-usee2bsandboxscreenshot
HarnessOpenShell DeepAgent
CodingSecurity
langchain-ai
General-purpose coding agent pairing LangChain Deep Agents with an NVIDIA OpenShell sandbox and Nemotron models
langchainsandboxnemotron
SWE-ReX
Deployment
SWE-agent
Sandboxed code execution for AI agents, locally or on the cloud. Massively parallel, easy to extend. Powering SWE-agent and more.
sandboxcode-executiondockerparallel-execution
ToolHermes Dynamic Workflows
Coding
lingjiuu
Hermes plugin for dynamic workflows — model-written sandboxed Python orchestrating up to 1000 subagents in parallel
multi-agentsandboxorchestrationparallel-execution
HarnessHermesClaw
SecurityDeployment
TheAiSingularity
Community stack running the Hermes Agent inside NVIDIA's OpenShell sandbox with kernel-enforced network and filesystem limits
sandboxdockeropenshellllama-cpp
PlatformVivaria
QAResearch
METR
Self-hostable web app, server and CLI for running agents on tasks in sandboxed environments and analysing the runs.
clisandboxtask-environments
Mistral Agents API
CodingConnectors
Hosted agents service with persistent conversations, MCP client support, and built-in connectors including sandboxed code execution.
mistralcode-executionsandboxconversation-state
Ona
CodingSecurity
Formerly Gitpod: API-first sandboxed cloud dev environments that host and orchestrate fleets of autonomous coding agents.
cloud-development-environmentsandboxcoding-agentgitpod
Osloq
QA
Agent that clones your repo into a sandbox, runs it and tries to reproduce GitHub issues, reporting evidence-backed results
githubsandboxissue-triagedebugging
Superconductor
Coding
Multiplayer workspace running many cloud coding agents (Claude Code, Codex, Pi, Amp, Cursor) in isolated sandboxes with shared sessions, previews and review.
coding-agentsandboxmulti-agentcode-review