SaaSHub helps you find the best software and product alternatives Learn more →
Top 23 agentic-ai Open-Source Projects
-
AutoGPT
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
AutoGPT (agpt.co): Open-source autonomous agent framework, released in 2023. Pioneering example of general-purpose agent architecture; known for demonstrating both promise and limitations of autonomous systems.
-
SaaSHub
SaaSHub - Software Alternatives and Reviews. SaaSHub helps you find the best software and product alternatives
-
git clone https://github.com/langgenius/dify.git cd dify/docker cp .env.example .env docker compose up -d
-
ragflow
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
Project mention: I Scanned 5 Popular Open-Source AI Projects for EU AI Act Compliance. Here's What I Found. | dev.to | 2026-03-31I ran AIR Blackbox (the scanner itself), Browser Use (79K+ stars), RAGFlow (76K+ stars), LiteLLM (23K+ stars), and Superlinked (15K+ stars) through the same compliance checks.
-
A free, structured 11-lesson course to start AI agents the right way. Think of it as turn-by-turn directions for agents, minus the detours and dead ends. Link: https://github.com/microsoft/ai-agents-for-beginners
-
ruflo
🌊 The leading agent meta-harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
Project mention: How I tracked down a 36GB memory leak in a Claude Code memory server | dev.to | 2026-06-21Upstream writeup with the full retainer trace: ruvnet/ruflo#2432. The wrapper itself, if you're curious: jazz-max/ruflo-hub.
-
anything-llm
Stop renting your intelligence. Own it with AnythingLLM. Everything you need for a powerful local-first agent experience
Project mention: NVIDIA RTX Spark: What the Backlash Gets Wrong About AI on Your Desktop [2026] | dev.to | 2026-06-04The headline marketing number is "1 petaflop" of AI performance. Sounds staggering. Tim Carambat, creator of AnythingLLM and one of the most credible voices in the local AI developer community, has already questioned this figure. His point is one I've validated repeatedly in my own benchmarking: for running large language models locally, memory bandwidth is the actual bottleneck, not raw FLOPS. You can have all the tensor cores in the world, but if you can't feed them data fast enough, your Llama 3 inference is still going to crawl.
-
Project mention: I Tested Flowise, Dify, and n8n Across 30+ Client Deployments. Here Is My Verdict. | dev.to | 2026-04-07
Citation Capsule: n8n's GitHub community reached 182,000+ stars across a 7-year development history, with 70+ AI-specific nodes added in 2024 to 2025. Source: n8n GitHub. Dify crossed 106,000 stars on GitHub with an Apache 2.0 license. Source: Dify GitHub. Flowise reached 51,000+ stars with MIT license. Source: Flowise GitHub. Dify's minimum recommended RAM is 4 GB versus Flowise's 1 GB and n8n's 300 MB. Source: Dify Docs.
-
OpenMontage
World's first open-source, agentic video production system. 12 pipelines, 52 tools, 500+ agent skills. Turn your AI coding assistant into a full video production studio.
Project mention: An open-source 'AI crew' that turns a coding assistant into a video studio | dev.to | 2026-07-01The project's goal, stated plainly on its project page, is to be the 'world's first open-source, agentic video production system.' Stripped of the buzzwords, that means two things. 'Open-source' means anyone can read, run, and change the code for free, rather than paying a company for a locked product. 'Agentic' means it is built around AI agents, software workers that do not just answer a single question but take on a goal, break it into steps, use tools, and carry the steps out with limited hand-holding. OpenMontage ships with twelve pipelines (start-to-finish recipes for common production jobs), fifty-two tools the agents can pick up and use, and more than five hundred skills, small bundles of instructions that teach an agent how to do a specific task well.
-
Birdclaw is interesting because it gives agents access to a Twitter archive. [17] GBrain points at a personal recall layer around OpenClaw / Hermes-style workflows. [18] PageIndex is a useful reminder that simple retrieval, even BM25-only retrieval, still has a place. [19] The “RAG comeback in about 8 months” take lands because the archive problem is still unsolved in practice. [20]
-
composio
Composio powers 1000+ toolkits, tool search, context management, authentication, and a sandboxed workbench to help you build AI agents that turn intent into action.
Composio provides managed authentication, per-user sessions, MCP access, SDKs, and a large catalog of pre-built tools. It is best suited for individual workflows and prototype agent builds where speed of setup matters more than centralized enterprise governance.
-
agenticSeek
Fully Local Manus AI. No APIs, No $200 monthly bills. Enjoy an autonomous agent that thinks, browses the web, and code for the sole cost of electricity. 🔔 Official updates only via twitter @Martin993886460 (Beware of fake account)
-
claude-skills
345 Claude Code skills & agent skills & plugins (30+ Agents, 70+ custom commands, 330+ skills, customizable references, scripts)for Claude Code, Codex, Gemini CLI, Cursor, and 8 more coding agents — engineering, marketing, product, compliance, C-level advisory, research, business operations, commercial & finance, and your daily productivity skills.
Project mention: I Built a Dual-Pool Adversarial Review System for AI Agents — And It Actually Works | dev.to | 2026-06-27I tested this on my own PR to alirezarezvani/claude-skills (18.7K stars):
-
agents-towards-production
End-to-end, code-first tutorials for building production-grade GenAI agents. From prototype to enterprise deployment.
Project mention: Show HN: GibRAM an in-memory ephemeral GraphRAG runtime for retrieval | news.ycombinator.com | 2026-01-17GraphRAG's approach certainly makes sense for regulation-heavy documents, where interconnected references can be critical for retrieval quality. The challenge you mentioned about separate systems for graph storage and vector indexing resonates with what we've seen in the field. In practice, this separation often adds unnecessary complexity for ephemeral analysis tasks. One thing we've explored is using a dual-memory architecture, which might be relevant here even in an in-memory context. It allows you to effectively manage recent context and longer-term semantic connections without the overhead of persistent storage. If you're interested in seeing how this approach could be integrated into production systems, I wrote up some patterns that might be useful: https://github.com/NirDiamant/agents-towards-production.
-
adk-python
An open-source, code-first Python toolkit for building, evaluating, and deploying sophisticated AI agents with flexibility and control.
The 497 open issues and 383 open PRs on the google/adk-python repository tell you something important: this is actively maintained, not another Google Labs experiment that'll get quietly sunset. With 3,600+ forks, people are building real things on top of it.
-
The memory interface shapes everything downstream. Committing to Hindsight's retain/recall abstraction early made the rest of the code straightforward. The LLM layer doesn't know where notes came from. The Flask layer doesn't know how retrieval works. Those boundaries are cheap at the start and expensive to add later.
-
keploy
Open-source platform for creating safe, isolated production sandboxes for API, integration, and E2E testing.
Project mention: Keploy – Traffic-Based Automated API Test Generator | news.ycombinator.com | 2026-03-31 -
Microsoft's agent-lightning project offers a comprehensive toolkit aimed at accelerating the process of building, testing, and deploying AI Agents. This open-source initiative highlights the industry's commitment to enabling faster development and implementation of advanced AI capabilities, providing developers with robust resources to streamline AI agent creation.
-
-
agent-framework
A framework for building, orchestrating and deploying AI agents and multi-agent workflows with support for Python and .NET.
Project mention: What We Learned Scanning Microsoft's Public Agent Framework Repository | dev.to | 2026-06-13Clear Code Intelligence scanned a public Microsoft repository: microsoft/agent-framework.
-
Project mention: HITL sdk VS humanlayer - a user suggested alternative | libhunt.com/r/u17g/hitl | 2026-06-17
HumanLayer is adjacent rather than a direct alternative: it focuses on human approval/contact flows for AI agents, while HITL SDK focuses on typed human-in-the-loop approval steps for durable workflow engines such as Temporal, Inngest, and Workflow SDK.
-
git submodule add https://github.com/The-Pocket/PocketFlow.git vendor/pocketflow
-
ART
Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.6, GPT-OSS, Llama, and more!
Huh?
I thought the latest advance in computing (spring 2025) is self-play / reinforcement learning. Like we've ran out of training data a few years ago.
https://github.com/OpenPipe/ART
Reinforcement learning having the large language model devise puzzles that they solve via llm-as-judge.
The definition of llm-as-judge is your llm generate 8-12 trajectories and a different llm judges the result. I'd use an oracle like windows or linux operating system execution for the problem of ISA-assembly creation.
The winning entries are used to train the large language model.
-
claude-code-tips
40+ tips for getting the most out of Claude Code, from basics to advanced - includes a custom status line script and Claude Code running itself in a container. Also includes the dx plugin: skills for everyday dev workflows.
NOTE:
The open source projects on this list are ordered by number of github stars.
The number of mentions indicates repo mentiontions in the last 12 Months or
since we started tracking (Dec 2020).
agentic-ai discussion
agentic-ai related posts
-
CorvinOS – self-hosted OS for AI agents, compliance built into the runtime
-
Talon: a self-hosted harness for long-lived AI agents
-
Talon: Self-hosted AI agent harness for chat, terminal, and desktop
-
CorvinOS – an agentic OS where GDPR/EU AI Act compliance can't be switched off
-
The first agentic reverse engineer
-
New AI tutor achieves 0.71-1.30 SD effect size in Dartmouth course [pdf]
-
CorvinOS – a self-hosted agentic OS which enforce EU AI Act 2026 by design
-
A note from our sponsor - SaaSHub
www.saashub.com | 11 Jul 2026
Index
What are some of the best open-source agentic-ai projects? This list will help you:
| # | Project | Stars |
|---|---|---|
| 1 | AutoGPT | 185,435 |
| 2 | dify | 148,169 |
| 3 | ragflow | 84,711 |
| 4 | ai-agents-for-beginners | 68,552 |
| 5 | ruflo | 63,860 |
| 6 | anything-llm | 63,062 |
| 7 | Flowise | 54,448 |
| 8 | OpenMontage | 36,269 |
| 9 | PageIndex | 33,921 |
| 10 | composio | 29,135 |
| 11 | agenticSeek | 26,609 |
| 12 | claude-skills | 21,882 |
| 13 | agents-towards-production | 20,939 |
| 14 | adk-python | 20,558 |
| 15 | hindsight | 18,172 |
| 16 | keploy | 17,967 |
| 17 | agent-lightning | 17,376 |
| 18 | memU | 13,980 |
| 19 | agent-framework | 12,019 |
| 20 | humanlayer | 11,083 |
| 21 | PocketFlow | 10,951 |
| 22 | ART | 10,381 |
| 23 | claude-code-tips | 9,098 |
Sponsored
SaaSHub - Software Alternatives and Reviews
SaaSHub helps you find the best software and product alternatives
www.saashub.com