self-hosted

AgentPort is a secure tool-calling layer for AI agents, with 50+ integrations, per-tool approval policies, encrypted credentials, and detailed call logs. It connects through MCP or CLI and is designed to let agents use external services without exposing secrets.

Self-hosted observability and control plane for production and coding AI agents, with live timelines, fleet-wide feeds, token budgets, MCP allow/block rules, and support for Claude Code plus a Python sensor.

Open-source DynamoDB-compatible API adapter with pluggable storage, built for local development, CI, self-hosted deployments, and air-gapped environments.

InsForge is an open-source backend platform for agentic coding, giving AI coding agents access to database, auth, storage, compute, hosting, and an AI gateway. The repo includes a clear quickstart, MCP server support, and both cloud and self-hosted workflows.

Crit is a local-first review tool for AI coding agents, letting users comment on plans and code diffs, iterate through multiple review rounds, and share reviews when needed. It supports several agent ecosystems and can be installed as a single binary.

A self-hosted Docker Compose stack for running local AI services including Ollama, LiteLLM, Whisper, Kokoro, embeddings, and an MCP Gateway.

An open-source mixture-of-agents server that fans prompts out to multiple LLMs, judges the responses, and synthesizes a final answer. It exposes an OpenAI-compatible API and also includes an agentic coding harness plus integrations for tools like Claude Code, aider, Continue, LangChain, and Pi.

Open-source memory SDK for AI agents with persistent cross-session memory, graph relationships, semantic search, and AI-powered memory processing. Offers both self-hosted and cloud modes, plus MCP support for tools like Claude Code.

MLJAR Studio is a private AI data analysis workspace that runs locally, helps users explore data in natural language, generates and runs Python, automates machine learning experiments, and turns notebooks into shareable web apps.

Cockpit is a web UI for Claude Code that lets you run and manage sessions from a browser instead of the terminal. It supports multiple concurrent sessions, transcript history, diff and file views, PR review workflows, scheduled runs, and settings for agents, skills, hooks, MCP servers, and CLAUDE.md.

A self-hostable SMS relay for AI agents that uses an old Android phone and a prepaid SIM, exposed over MCP so clients like Claude, ChatGPT, and Cursor can send and receive real texts.

RiskKernel is an open-source, self-hosted reliability runtime for AI agents. It adds hard budgets, kill switches, crash-resumable runs, human approval gates, and OpenTelemetry export, with support for OpenAI, Anthropic, and existing agent stacks.

Wyolet Relay is a self-hosted LLM router that exposes OpenAI- and Anthropic-compatible endpoints in front of multiple providers. It pools API keys, handles failover and rate limits, tracks usage and cost, and includes a quickstart plus Docker-based deployment.

Self-hosted runtime for putting hard budgets, approval gates, observability, and resumable checkpoints around AI agents.

Open-source, self-hostable sandbox for AI agents. It lets agents work inside isolated Linux containers through MCP tools, with CLI and host-level admin controls for creating, managing, and exposing boxes.

MakerChecker is a self-hosted governance layer for AI agents that enforces roles, approvals, and tamper-evident audit logs. The repo shows quickstart commands, a demo flow, and SDK/connectors for TypeScript, Python, and LangChain-style integrations.

Open-source agentic end-to-end browser testing tool for web apps. It runs JSON-defined test journeys in Docker and can be used as a CLI or optional self-hosted web app.

Open-source interface for running coding agents from phone, desktop, web, and CLI. Supports multiple providers, runs locally on your own machine, and includes a daemon plus client apps.

Self-hostable chat network for AI agents with rooms, DMs, persistent history, and pre-built bridges for Claude Code, Codex, and Claws runtimes.

SwarmWright is a self-hosted multi-agent AI orchestration platform for designing, running, and inspecting swarms of agents from markdown and config. The page highlights Docker-based setup, topology-enforced agent graphs, human approval steps, local storage, and detailed audit logs.

Self-hostable MCP server that collects memories from coding agents and regenerates shared AGENTS.md-style context for the next session.

Open-source agile planning and orchestration layer for AI coding agents, with bidirectional sync to Claude Code, AI-generated user stories, sprint planning, and a self-hosted CLI workflow.

Open-source debugging agent for developers that connects coding agents to production runtime data to help identify bugs, triage issues, and generate fixes. The repo also includes a CLI and web dashboard for reviewing sessions and turning investigations into branches or pull requests.

Local-first encrypted memory engine for AI agents, built in Rust with MCP support and persistent cross-session memory for Claude and other MCP clients.