Overview
Linchpin is an open-source, self-hostable runtime for AI agents. It lets you deploy agent environments on your own infrastructure using a single Docker Compose command. Users can run agents that connect to any cloud model via OpenRouter (supporting ~200 models) or local models through Ollama. Each agent session runs in a sandboxed Docker container with preinstalled tools like Python, Node, git, and ripgrep. The platform is designed for developers who need secure, isolated, and flexible agent execution.
Application scenarios
Secure code execution
Run untrusted code from AI agents in isolated Docker containers with per-environment networking controls.
Multi-model agent workflows
Switch between cloud models (Claude, GPT, Gemini, Llama) and local models per agent without changing infrastructure.
Live UI streaming
Subscribe to session event logs over SSE for real-time monitoring or building live agent interfaces.
Credential management
Store and inject API keys or secrets securely into agent sessions without exposing plaintext data.
Tool integration
Plug in external tools via MCP servers or HTTP endpoints to extend agent capabilities.
Core features
Any model, one adapter
OpenRouter routes to ~200 cloud models (Claude, GPT, Gemini, Llama, DeepSeek, Mistral, Qwen), while Ollama runs any local model you've pulled. You can switch providers per agent.
Sandboxed sessions
Every session gets its own Docker container with Python, Node, git, and ripgrep preinstalled. Networking is per-environment—none for tight isolation, unrestricted for open egress.
Eight built-in tools
bash, read, write, edit, glob, grep, web_fetch, and web_search run inside the container, so the model can't escape.
MCP & HTTP tools
Plug in Model Context Protocol servers via stdio or point at any HTTP endpoint. The connector handles process lifecycle and credential injection.
Credential vaults
Fernet-encrypted credential store. Reference secrets by name from agent configs; they decrypt at session start and never hit disk in plaintext.
Event streaming
Append-only event log per session with cursor pagination. Subscribe over SSE—replays anything past your cursor, then streams live. Perfect for live UIs.
Any model, one adapter OpenRouter routes to ~200 cloud models (Claude, GPT, Gemi
Any model, one adapter OpenRouter routes to ~200 cloud models (Claude, GPT, Gemini, Llama, DeepSeek, Mistral, Qwen), while Ollama runs any local model you've pulled. You can switch providers per agent.
Sandboxed sessions Every session gets its own Docker container with Python, Node
Sandboxed sessions Every session gets its own Docker container with Python, Node, git, and ripgrep preinstalled. Networking is per-environment—none for tight isolation, unrestricted for open egress.
Target users
Developers and teams building or deploying AI agents who need self-hosted, secure, and flexible runtime environments. This includes backend engineers, ML engineers, and DevOps professionals who want to control infrastructure, isolate agent execution, and integrate with multiple model providers.
How to use
Deploy Linchpin with a single Docker Compose command on your own server. After deployment, configure agents to use any cloud model via OpenRouter or local models via Ollama. Set up credential vaults to store secrets, define per-session networking rules, and connect built-in tools or external MCP/HTTP endpoints. Monitor sessions via the event stream with cursor pagination and SSE for live updates.
Effect review
Linchpin delivers a practical, developer-focused runtime for AI agents with strong security and flexibility. The sandboxed sessions and credential vaults address real concerns about running untrusted code and managing secrets. The support for both cloud and local models, along with the ability to switch providers per agent, gives teams significant freedom. The event streaming feature is a smart addition for building live UIs. For developers who need a self-hosted agent runtime, Linchpin offers a clean, well-designed solution without unnecessary complexity.
Frequently asked questions
What is Linchpin?
Linchpin is an open-source, self-hostable runtime for AI agents that supports any cloud model via OpenRouter or local models through Ollama, deployable with a single Docker Compose command.
How do I deploy Linchpin?
You can deploy Linchpin with a single Docker Compose command, making it easy to set up and run on your own infrastructure.
What AI models does Linchpin support?
Linchpin supports any cloud model via OpenRouter and local models through Ollama, giving you flexibility in choosing the AI backend.
Is Linchpin free to use?
Yes, Linchpin is open-source and free to self-host. You only pay for the AI models you use through OpenRouter or local resources.
Can I run Linchpin locally without internet?
Yes, by using Ollama with local models, Linchpin can run entirely offline on your own hardware.
Launch URL
https://linchpin.work/Tags
Featured recommendations
OpenCode
alternativeOpenCode by OpenCode.ai is an AI-powered development platform for generating, reviewing, and debugging code, enabling faster software creation and collaboration.
LobeHub
alternativeLobeHub provides a platform to find, build, and collaborate with AI agent teammates for both work and life, designed to grow with you.
Craft Agents
alternativeAn open-source desktop app by Craft for working with AI agents. Connect any API or MCP server, multitask naturally, and customize workflows.
OpenFang
alternativeOpen-source Agent OS by OpenFang, built in Rust. Features 7 autonomous Hands, 53 tools, and 16 security systems for automated workflows, with support for multiple LLM providers and a WASM sandbox.
Liner
alternativeLiner by Liner Corporation is an AI tool that helps professionals search with sources, research academic papers, and write clearly in one workflow.
Paseo
alternativeA self-hosted daemon by Paseo for Claude Code, Codex, and OpenCode. Agents run on your machine with your full dev environment, accessible from phone, desktop, or web.
Kilo
alternativeKilo's AI agent platform for deploying OpenClaw, offering access to 500+ models with enterprise-grade security and requiring no DevOps setup.
Moltis
alternativeMoltis is a secure, Rust-native AI assistant that runs on your hardware. It offers sandboxed execution, multi-LLM support, voice capabilities, memory, and integrates with platforms like Telegram, What
Related Toolkits
AI Assistant / AgentsCraft Agents
An open-source desktop app by Craft for working with AI agents. Connect any API or MCP server, multitask naturally, and customize workflows.
View Details
AI Assistant / AgentsBrowserOS
A free, open-source AI Agent browser that integrates powerful AI agents directly into the browser.
View Details
AI Assistant / AgentsChuangyi AI
An AI multi-agent collaborative creation platform that provides AI employee outsourcing services for short film and short drama creators, unleashing content production efficiency!
View Details