Self-host GoGogot
v0.1.1Lightweight self-hosted AI agent in Go
- Stars
- 263
- CPU
- any
- RAM
- ~10 MB
- Storage
- minimal
octagonlab/gogogotWhat is GoGogot?
GoGogot is a lightweight, self-hosted AI agent written in Go (~8,800 lines) that lives on your server and communicates via Telegram. It provides 34 built-in tools including shell command execution, file operations, web search and fetching, persistent memory and skills, identity management, task scheduling, and conversation recall. The agent runs as a single ~15 MB binary using ~10 MB RAM at idle, deployed via Docker or Docker Compose. It supports Anthropic, OpenAI, and OpenRouter models with your own API keys. All data (conversations, memory, skills, identity files) stays on your mounted volumes as plain files. The agent polls Telegram for messages and makes outbound calls to LLM providers — it opens no inbound ports. Authentication is via a single Telegram owner ID. It is MIT licensed and has no paid tiers.
Should you self-host GoGogot?
GoGogot scores 42/100 on selfhost.directory (how scoring works) and ranks #79 of 120 in AI Assistant. It has 263 GitHub stars.
It shipped 11 stable releases in the past year. The latest, v0.1.1, came out on 28 October 2026. New versions land roughly every 1 days.
Setup is moderate: expect some configuration beyond a single command. An official Docker image is available. It is MIT-licensed and written mostly in Go.
Its closest alternative on selfhost.directory is hermes-agent (89/100, 257 885 stars). It scores higher overall, so compare the two before you pick. See GoGogot vs hermes-agent side by side.
Features
- Context Compaction
- Conversation recallUses LLM to select relevant past chats, not vector embeddings
- Extensible Architecture
- File operations
- Identity Management
- Multi-Model Support
- No inbound ports
- Persistent memory
- Shell command execution
- Skills System
- Task Scheduling
- Telegram Interface
- Web search and fetchweb_search requires BRAVE_API_KEY
GoGogot README
More in AI Assistant
golembotGolemBot connects coding agents (Cursor, Claude Code, OpenCode, Codex) to Slack, Telegram, Discord and more — routing through any LLM provider via one config file.
GPTPortalGPTPortal is a self-hosted, open-source Node.js web app providing a unified chat interface for multiple LLM providers and local endpoints like Ollama.
hal0hal0 is a self-hosted, open-source AI inference platform turning a Linux machine into a private OpenAI-compatible appliance.
Hass-MCPA self-hosted Model Context Protocol server that lets AI assistants like Claude query and control Home Assistant entities, automations, and dashboards.
hermes-agentSelf-hosted open-source AI agent that learns from experience, creates skills, runs on Telegram, Discord, Slack, CLI — MIT licensed.
hermes-agent-with-webuiHermes WebUI is a self-hosted, open-source web interface for the Hermes Agent with persistent memory, scheduled jobs, and multi-device access. Runs on Python 3.11+.
Hermes CompanionSelf-hosted desktop companion for Hermes Android enabling mesh task dispatch and conversation handoff over local network with end-to-end encryption.
Hermes-RelayHermes-Relay is an open-source, self-hosted Android app and desktop CLI that connects to your own Hermes AI agent for streaming chat, voice, and remote terminal access.
hermes-uiSelf-hosted web UI for Hermes Agent AI assistant. Single HTML file with Python proxy, featuring chat, file browser, task board, skills, MCP tools, and three themes.
Hermes-WorkspaceHermes Workspace is a self-hosted, open-source web UI command center for AI agents — combining chat, terminal, memory, skills, and multi-agent orchestration.
hivekeepSelf-hosted AI agent team with persistent memory, collaboration, custom tools, and 6 messaging channels — all in one container.
HomeGenieSelf-hosted, open-source home automation server for Windows, Linux, macOS and Raspberry Pi, adding local LLM agents, AI vision, voice control and scriptable automations.
0 comments