Resource Hub
3238 hand-picked resources, updated every week. Search it, filter it, or just browse a collection and see what catches your eye. Want today’s headlines instead? Read the free AI news feed.
Filtered by tag
Answers come only from resources in this hub, with the sources listed.
42 resources
An OpenAI Daybreak initiative, with Trail of Bits, Calif and HackerOne, that validates AI-found vulnerabilities in open-source projects, tests patches, and coordinates disclosure with maintainers. Projects can apply to take part.
Google Open Source's glossary of internal tool and process names, such as google3, Critique, Changelist and Forge, that appear in Google's open-source projects.
An open-source managed agent platform on GitHub, with an API for running AI agents.
Open-source, Kubernetes-native framework for running large language model inference across many machines at high performance.
MakeUseOf first-person review of Vane, an open-source, self-hostable AI answering engine positioned as an alternative to Perplexity.
Open-source AI-powered answering engine that searches the web and answers questions with cited sources; can be self-hosted.
The MIT-licensed open-source repository for REA, a tool that lets coding agents reverse engineer software, from app behavior down to native binaries.
An ExplainX roundup of ten public GitHub projects that credit the REA reverse-engineering tool, such as UI kits and apps rebuilt from existing software, with links to each repository.
NVIDIA's organization page on Hugging Face, hosting publicly listed AI models, datasets, and related resources, including the Nemotron dataset family. Individual assets carry their own license terms.
MIT-licensed Git platform by Jovi De Croock for plan-driven development, agent-assisted review, merge queues and controlled rollout. The repository provides source and deployment instructions. Software access is free; hosting and any model services used may cost money.
Free, open-source Git hook that pauses protected-branch pushes to ask questions about what the changed code does. Supports plain Git and several coding-agent workflows. The software is free; use of an external model endpoint may incur separate costs.
Guide to an MIT-licensed collection of Claude Code skills for analyzing and rebuilding app workflows, from reconnaissance through testing and deployment. Discusses legal, quality and security caveats; the skills are free, but model access is separate.
Substack post releasing agent-engineer, an open-source plugin for Claude Code, Codex, and Cursor that adds senior engineering skills, subagents, and hooks to AI coding assistants.
Open-source, vendor-neutral observability framework for cloud-native software. It provides a single set of APIs, libraries, agents, and collector services to capture distributed traces, metrics, and logs, built on the earlier OpenTracing and OpenCensus projects.
The software page of the Berkeley AI Research (BAIR) lab at UC Berkeley, listing the lab's open-source projects with links to source code. Projects include Caffe, the deep learning framework, and Ray, the distributed computing framework with libraries for hyperparameter search, reinforcement learning and model serving.
An open benchmark for testing how well large language models resist jailbreak attacks — prompts designed to get a model to produce harmful or unwanted content. It provides a public repository of jailbreak prompts, a standardized evaluation library with a defined threat model and scoring, and a leaderboard of attacks and defenses.
Documentation for StrongREJECT, an open-source benchmark and Python package for evaluating LLM jailbreaks. It includes rubric-based and fine-tuned evaluators, several dozen baseline jailbreaks, and a dataset of prompts across six categories of harmful behavior, from disinformation to violence.
A plain-language guide from OpenK3 explaining what open-weight and closed-weight AI models are, how licenses still restrict open-weight models, real examples including Kimi K3, and how teams can choose between self-hosting and using a hosted API.
Open-source runtime from NVIDIA for running autonomous AI agents in a sandboxed, private environment. Code and documentation are on GitHub.
Open-source tool that routes each Claude Code task to the lowest-cost model suited to it, to reduce token spending.
Nonprofit building open-source tools that let people get insights from data they cannot directly access or share.
Open-source, low-code visual builder for agentic and retrieval-augmented generation (RAG) AI applications. Free to self-host; the project also offers hosted options.
A free directory that highlights open-source GitHub repositories and developer tools, with short write-ups sorted by language and topic.
Microsoft's open-source web framework for building web apps and services in C# that run on Windows, Linux and macOS. Free to download and use; the page is Microsoft's own product page.
Open-source personal AI assistant that runs locally on your own devices and connects to WhatsApp, Telegram, and Discord. Features persistent memory, proactive notifications, and community-built skill extensions. Formerly known as ClawdBot. Free and open source.
The most widely used open-source framework for building and training AI models, with free tutorials, documentation and pre-trained models. Most AI research papers today are built on it.
Why I recommend it: Free and open source, now governed by the Linux Foundation's PyTorch Foundation — but Meta created it and remains its biggest contributor, so its direction still reflects big-tech priorities.
Google's open-source framework for building and training AI models, with free guides and pre-trained models. Once the industry standard, it's now used less in new research than PyTorch but still powers many production systems.
Why I recommend it: Free and open source, but it's a Google project — its development priorities follow Google's. For a new learner, PyTorch is the more common starting point in 2026.
PrettyTable is a free, open-source Python library for drawing formatted ASCII tables in a terminal or text output. This post from Open-source Projects shows what it does and links to the GitHub repo.
Why I recommend it: Free open-source library (MIT license). The link goes to a blog post on opensourceprojects.dev; the project itself lives at github.com/prettytable/prettytable and pypi.org/project/prettytable. The blog runs on ads and sponsorships.
Community home of modern Fortran, with free compilers, tutorials and a package manager.
Why I recommend it: Free and community-run. Still widely used in weather, physics and engineering research.
Official Perl site with free downloads, documentation and learning links for this long-running scripting language.
Why I recommend it: Free and open source. Mostly found in older systems today — worth learning if a job posting asks for it.
Official home of Julia, a free, open-source language built for fast numerical and scientific computing, with docs and tutorials.
Why I recommend it: Completely free and open source. Useful if you work with math-heavy data; the job market is smaller than Python's.
Open-source skill that records video demos of your web app with zoom-ins and voiceover.
Why I recommend it: Free on GitHub. Needs some technical setup; I haven't tested it myself.
Bespoke Labs' open-source toolkit for local typed decisions, contrastive data curation and model evaluation.
Why I recommend it: Free and open source, from a company that also sells services — the repo is usable on its own.
Open-source "System 1" decision engine: typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.
Why I recommend it: Free and open source, but an individual developer's project — check the commit history and issue activity before relying on it for anything important.
A Jev-like family of small decision models built on Qwen that you can train and run on your own hardware, from Jared Palmer.
Why I recommend it: Free and open source. Running it yourself means paying for your own compute; the models are small enough for a laptop or a cheap server.
Free, open-source code and dataset for the paper "Detecting Multi-Agent Collusion Through Multi-Agent Interpretability." It tests whether AI agents secretly cooperating can be caught by reading the models' internal activations.
Why I recommend it: A research tool, not a beginner resource. Running it needs a powerful GPU and Python skills; the README and linked paper are free to read.
Free question-and-answer site explaining AI risk arguments in plain language, founded by Rob Miles and maintained by volunteers. Answers are organised as linked questions from beginner to advanced, covering how AI is advancing, why systems may pursue goals, alignment research and AI governance. Includes Stampy, a chatbot that answers AI safety questions with sources. Open source on GitHub; run as a project of Ashgro Inc, a US 501(c)(3) charity.
Why I recommend it: The clearest free place to find out what people mean when they talk about AI risk, written so you can follow it without a technical background. Be clear about what it is: this is advocacy, not a neutral survey of the debate. The homepage opens with 'it could lead to human extinction', and the whole site is built by people who already hold that view, so you will get their strongest arguments rather than the strongest objections to them. Their own chatbot warns it can be inaccurate — check its sources before repeating anything. Read it to understand the case, then read the critics of it, and pair it with the AI Basics page here for the numbers.
A full AI assistant — writing, images, web search, memory, file uploads — where every conversation is end-to-end encrypted on your device, so the company says it cannot read your chats, train on them, hand them to partners, or produce anything but scrambled text in response to a subpoena. Built by Moxie Marlinspike, the cryptographer who created Signal. Free to start with no credit card; the encryption and private inference designs are written up publicly and the code is open source so the claims can be checked.
Why I recommend it: This is the one to reach for when you are about to type something into an AI that you would not want read back to you — money trouble, health, a manager, a visa problem. Two honest things. It is free to start, which is not the same as free forever, so read the plan page before you rely on it. And encryption protects the message, not your judgment: anything you paste in that belongs to an employer or a client is still their information, whoever can or cannot read it.
Unified intelligence platform that turns structured and unstructured data into a governed knowledge graph for AI. Offers a free open-source graph database (FlureeDB) and a hosted Fluree AI tier that starts at $0 with a free fuel allowance; paid enterprise plans add scale, SSO and private deployments.
From the site: Fluree turns raw data into trusted, queryable knowledge graphs. GraphRAG-powered accuracy for enterprise AI.
The intelligent orchestration platform for DevSecOps, enabling teams and agents to ship trusted software at enterprise scale.
From the site: The intelligent orchestration platform for DevSecOps, enabling teams and agents to ship trusted software at enterprise scale.
Why I recommend it: Free tier includes unlimited public and private repositories with core CI/CD minutes; paid plans add enterprise features and more minutes.
Open-source workspace from Block where a team and its AI agents share the same room — chat, planning, project tracking, code and pull requests in one place, with agents you configure for your own workflows. Currently a developer preview; the app is open source on GitHub, with a waitlist for a paid enterprise version.
From the site: Come test the early stages with us.
Why I recommend it: Early-stage, so treat it as something to try rather than to run a team on. The interesting part is watching how human-plus-agent teamwork gets designed — useful reading if you want to talk credibly about agents at work. The open-source app is free to run yourself; the enterprise version will be paid, and there is no published price yet.
Public leaderboard and open-source benchmark that drops AI agents into realistic business environments with 47 real tools across sales, marketing, operations, support, finance, and HR. Scores are based on final environment state, not an LLM-as-judge.
Why I recommend it: The leaderboard and the benchmark code are free; running it yourself means paying the model APIs at the costs shown. The test design is based on Zapier's own task data, so it's a realistic lens on agent work, but Zapier also sells automation tools — treat the benchmark as a useful public dataset, not a neutral referee.
