Open Source AI Agents
Discover 406 open source AI Agents repositories from GitHub, each with an analysis of what it does, key features, use cases and alternatives. AI Agents projects here are most often combined with Python, Developer Tools and MCP. Last updated October 4, 2026.
406 repositories · updated October 4, 2026
deliberation: Get Multi-Model Reviews and Consensus
Deliberation is an MCP server and coding-agent plugin that routes plans and code questions to external AI models for second opinions or arbiter-mediated consensus. It suits developers who want independent reviews before making consequential changes.
cabinet: Build a Self-Hosted AI Knowledge Base
Cabinet combines a local, Markdown-based knowledge base with AI agents, scheduled jobs, and team workflows. It is aimed at people who want persistent AI context and automation while keeping their files on disk and under their control.
openclaw: Run a Personal AI Assistant Across Your Devices
OpenClaw runs an AI assistant on your own hardware and connects it to messaging channels and companion apps. It suits people who want a customizable assistant across devices, with local control over its state and configuration.
superpowers: Guide Coding Agents Through Software Development
Superpowers adds a structured development methodology to coding agents, from clarifying requirements and planning through test-driven implementation and review. It suits developers who want repeatable workflows across supported agent tools.
zeroboot: Create Fast VM Sandboxes for AI Agents
Zeroboot uses copy-on-write forks of Firecracker VM snapshots to create isolated sandboxes quickly for code execution. It is aimed at teams running agent-generated code that need VM-level isolation and can accept a working-prototype maturity level.
compound-engineering-plugin: Guide AI Agents Through Engineering Work
Compound Engineering provides reusable skills for AI coding agents, guiding work from brainstorming and planning through implementation, review, and captured learnings. It supports multiple agent hosts, including Claude Code, Cursor, and Codex.
feynman: Research Papers with an AI Agent
Feynman is a TypeScript CLI agent for investigating research topics, reviewing papers, and comparing sources. It suits researchers and developers who want structured, citation-aware literature work and can configure a model provider.
mcp-grafana: Connect AI Clients to Grafana
Grafana’s MCP server lets compatible AI clients access Grafana dashboards, data sources, alerts, incidents, and other stack features through the Model Context Protocol. It suits teams that want to bring Grafana context and selected operations into agent workflows.
engram: Give AI Coding Agents Persistent Memory
Engram is a Go-based memory system that helps AI coding agents retain useful project knowledge between sessions. It stores memories locally in SQLite and makes them available through MCP, a CLI, an HTTP API, and a terminal UI.
hermes-webui: Use Hermes Agent in a Web Browser
Hermes WebUI provides a self-hosted browser interface for Hermes Agent, with chat, session management, and workspace browsing. It suits people who already run Hermes and want to access it from a desktop or phone without using the terminal.
GitNexus: Map Codebases for AI-Assisted Analysis
GitNexus indexes a codebase as a knowledge graph and exposes its relationships through CLI and MCP tools. It helps developers and AI coding agents trace dependencies, understand execution flows, and assess change impact.
autoresearch: Let AI Agents Run LLM Training Experiments
autoresearch gives an AI coding agent a small, single-GPU language-model training setup to modify and test autonomously. It suits researchers with supported NVIDIA hardware who want to explore training changes through repeated, time-bounded experiments.