Self-hosted personal AI agent with a built-in WebUI, memory, tools, MCP support, sub-agents and scheduled automations, using local or cloud LLMs.
pardnchiu/Agenvoy
A self-hosted Go AI agent that runs 24/7 on your own computer, auto-builds missing tools, and shares them with Claude Code and other agents via MCP.
Open-source alternative to:OpenAI ChatGPT
- Stars
- 559
- Forks
- 54
- Open issues
- 3
- Last commit
- 2 days ago
What is it?
Agenvoy is a self-hosted AI agent harness written in Go. It runs as a single binary on your own machine and turns requests into completed work: it searches live data, works with local files, runs multi-step tasks, and creates schedules that push results back to you. Memory, schedules, tools and credentials stay on the local machine.
When no suitable tool exists, the agent builds, tests and saves one for reuse. Tools run in a sandbox, and access to sensitive paths requires confirmation. Agenvoy also acts as an MCP server, so Claude Code, Codex and other MCP-compatible agents can use the same tool library, and it can connect to external stdio or HTTP MCP servers as a client.
A browser dashboard served locally manages sessions, tools, schedules and memory. Telegram and Discord are supported as chat channels through outbound connections, so no inbound ports need to be opened. Models can be routed by task, and Ollama is supported for local or cloud models.
Key features
- 01Runs as a single Go binary, 24/7, on your own machine
- 02Generates, tests and saves missing tools automatically
- 03Acts as an MCP server and client, with stdio and HTTP transports
- 04Creates recurring schedules from a single sentence
- 05Runs tool commands in a sandbox with confirmation for sensitive paths
- 06Connects to Telegram and Discord without opening inbound ports
Use cases
- Running a daily research or reporting job that pushes results to chat
- Searching and answering questions over local documents
- Letting Claude Code or Codex reuse the same sandboxed tools
- Accessing a private local agent from Telegram or Discord
Things to know
- Licensed under AGPL-3.0, a copyleft license that requires derivative network services to share their source code.
- Runs as a self-hosted daemon on your own computer, so you must set up and keep that machine running 24/7.
In collections
Topics
FAQ
- What is Agenvoy?
- A self-hosted Go AI agent that runs 24/7 on your own computer, auto-builds missing tools, and shares them with Claude Code and other agents via MCP. Agenvoy is a self-hosted AI agent harness written in Go. It runs as a single binary on your own machine and turns requests into completed work: it searches live data, works with local files, runs multi-step tasks, and creates schedules that push results back to you. Memory, schedules, tools and credentials stay on the local machine.
- Is Agenvoy free to use? What license does it have?
- Agenvoy is open source under the AGPL-3.0 license. Check the license terms before using it in commercial products.
- Is Agenvoy still maintained?
- Actively maintained. The last commit to Agenvoy was on Nov 5, 2026.
- What can I use Agenvoy for?
- Running a daily research or reporting job that pushes results to chat Searching and answering questions over local documents Letting Claude Code or Codex reuse the same sandboxed tools Accessing a private local agent from Telegram or Discord
- Where can I get Agenvoy?
- The source code and installation instructions are on GitHub: https://github.com/pardnchiu/Agenvoy
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
A self-improving AI agent that creates and refines skills from experience, runs on many model providers, and works through CLI and messaging apps.
Open-source desktop AI office assistant that plans and carries out file, document and command tasks on your machine from a one-sentence request.
Self-hosted workspace for creating named AI coworkers with their own instructions, memory, routines, and approval controls.
A personal AI assistant that runs on your own computer, handling long tasks, remembering context, and talking by voice with any local or cloud model.
A self-hosted, provider-agnostic web interface for chatting with local Ollama models and OpenAI-compatible APIs, with RAG, plugins and multi-user access control.