Explore the directory

A hands-on workshop sample that builds a retrieval-augmented-generation chat application with LangChain.js and OpenAI, deployable to Azure Static Web Apps, Azure Container Apps and Azure AI Search.

• RAG
• Tutorials

A curated list of over 500 community plugins for DeepSeek Harness, an open-source agent runtime in which the coding agent, its tools, models, sandbox, session storage, UI and agent loop are all separately replaceable plugins.

• Agent skills
• Curated lists

A .NET application and library for generating code from FHIR healthcare-interoperability specifications, including cross-version mapping and multiple output languages.

• Libraries
• APIs

An agent-native command-line development tool: an AI coding assistant continuously fine-tuned through community-driven incentives, focused on optimizing code for performance and system efficiency.

• Coding agents

A Claude skill that interviews you before any code is written and generates a shared requirements, design and task plan, then creates matching configuration files so Claude, Cursor and Copilot stop contradicting each other.

• Agent skills

An open-source AI video-editing plugin and local MCP server that turns a folder of raw clips and a plain-language description into a finished, beat-synced cut, driven from Claude Code, Codex, Hermes or OpenCode.

• MCP servers
• Generative media

A human-generated, multi-turn benchmark and four document corpora for evaluating retrieval-augmented-generation systems on realistic, multi-turn conversations.

• RAG
• AI evals

A unified manager and router for llama.cpp, MLX and vLLM model instances, with a web dashboard, a built-in model downloader, and OpenAI- and Anthropic-compatible routing across multiple running models.

• Inference

A full-stack tutorial project that turns a topic, a block of text, or an uploaded document into an editable slide outline, then generates and exports a native, still-editable PowerPoint file.

• Tutorials

A Swift library for running text-embedding models locally using Apple's MLTensor, supporting BERT, ModernBERT, NomicBERT, RoBERTa, XLM-RoBERTa and Qwen3 embedding architectures loaded directly from Hugging Face.

• RAG
• Inference

An MCP server that gives Claude, Cursor and other AI coding tools persistent session memory across restarts, routing simple tool calls to a bundled local model fleet instead of the cloud and reporting the tokens saved.

• MCP servers
• Agent memory

A large collection of 1297 interview questions and answers on Retrieval-Augmented Generation, covering 52 RAG architectures, system design scenarios, and production failure modes for AI/ML engineers.

• RAG
• Tutorials

A high-performance vector database using 1-bit quantization and hyperbolic geometry to model hierarchical data, aimed at autonomous agents and robotics rather than static document search.

• Agent memory
• Databases

An offline Windows desktop app for organizing, searching and copying prompts, storing them in a local SQLite database with no browser, account or network connection.

• Desktop utilities

A minimal Model Context Protocol server that gives an LLM a single tool for evaluating a mathematical expression precisely, instead of relying on the model's own arithmetic.

• MCP servers

A terminal-native AI coding assistant built on DeepAgents CLI that inspects and edits project files, runs approved shell commands, keeps long-term memory across sessions, and can bridge scheduled or remote tasks in from WeCom messages.

• Coding agents

A self-hosted AI assistant runtime meant to run real business workflows, such as CRM, marketing and finance tasks, reliably on smaller local models, with credentials kept out of the model's context and every action gated by approval tiers.

• Agent frameworks
• Workflow automation

balena's on-device agent for IoT hardware, responsible for installing, starting, stopping and monitoring OCI-compliant containers and managing an application's lifecycle on the device.

• Containers
• Hosting

A research framework for automatically creating specialized sub-agents at runtime, modeling each as a four-part interface of instructions, context, tools and model rather than a fixed, hand-engineered role.

• Agent frameworks

An on-device, typed property graph database with a schema defined as version-controlled code, distributed as a single CLI with no server to run.

• Databases

A reinforcement-learning post-training framework that keeps trillion-parameter language models at low-precision deployment format and trains only a small BF16 adapter, fitting frontier-scale RL onto a single node.

• Models

A local-first native desktop AI workspace bundling multi-provider chat, a coding agent with diff-based file edits, and other agent-driven workspaces such as an ERP builder and a 3D model generator, with no backend or telemetry.

• Coding agents
• Inference

A Model Context Protocol server that lets Claude and other MCP clients list, search, create, update, delete and clear tasks in a user's Google Tasks account.

• MCP servers

A daily-updated, automatically generated list of every currently free LLM API, ranked by a quality score and checked for live availability, pulled from OpenRouter, Pollinations AI and other public sources.

• Curated lists