Explore the directory

A set of official SAPUI5 and OpenUI5 plugins that help coding agents create new UI5 projects, detect and fix UI5-specific errors, and get up-to-date UI5 reference information while working in a project.

• Agent skills

A research technique that attaches a small hypernetwork to a recursive transformer block, generating a different LoRA modulation at each recurrence step instead of repeating the same transformation, improving depth-recycled models.

• Models

An alpha-stage terminal application that manages a kanban board of tickets and launches, monitors and pauses multiple AI coding agents against them across many project repositories from one dashboard.

• Coding agents

An open-source terminal coding agent that reads a project's files, writes code, runs tests and fixes failures, working with hosted providers through the user's own API key or with fully offline local models.

• Coding agents

An ACL paper's official code for a logical data-poisoning attack on GraphRAG systems, which swaps entities of the same type to corrupt a corpus's implicit reasoning graph while keeping the poisoned text locally plausible.

• Security
• Papers

A curated set of category pages linking to ChatGPT and LLM applications, prompt engineering guides, agent frameworks, coding tools, libraries and other adjacent developer resources.

• Curated lists

A prompt and methodology repository for extracting a three-layer knowledge model, scenarios, concepts and entities, from raw articles, then answering questions strictly from that modeled knowledge with sources cited.

• Agent skills
• Agent memory

An end-to-end LLM toolkit that runs entirely on one consumer NVIDIA GPU, taking a user from environment diagnosis through pretraining, fine-tuning, evaluation, GGUF export and a self-contained benchmark report.

• Models
• Tutorials

A PyTorch implementation, from an ACL paper, of knowledge-graph retrieval-augmented generation for LLM-based recommendation, evaluated on MovieLens and Amazon Book datasets.

• RAG
• Papers

A single-binary, Go implementation of a RAG workspace and autonomous agent framework, inspired by the Leann backend, that runs embeddings and inference locally through Ollama or llama.cpp with AST-aware code indexing.

• Agent frameworks
• RAG

A self-hosted, three-tiered memory system for AI agents that stores and semantically searches memories through a bundled Qdrant and FastEmbed container, exposed over MCP with no API keys or cloud dependency.

• MCP servers
• Agent memory

An Android SDK that gives large language models typed, structured access to phone capabilities such as calendar, contacts, SMS and sensors through the Model Context Protocol, with everything staying on the device.

• MCP servers
• Mobile

A single-binary local LLM inference platform, inspired by Ollama, that downloads and runs models from the CSGHub platform with a web UI, an OpenAI-compatible API, resumable downloads and one-click coding-agent setup.

• Inference

An automated pipeline that converts Hugging Face models to GGUF format and uploads the results, running weekly over every model in the ggml-org organization.

• Models

An LLM-powered tool that turns a natural-language causal question and a dataset into a full causal-inference analysis, selecting a suitable method, running it, checking diagnostics and explaining the result in plain language.

• Agent frameworks
• Data engineering

A TypeScript-native, headless spreadsheet-workbook runtime for Node.js services that edits inputs, recalculates formulas, reads outputs and persists WorkPaper JSON, with MCP tools for AI agents.

• MCP servers
• Libraries

An autonomous agentic system that develops supervised machine-learning models for biomedical data, running experiments in a sandboxed Docker container and producing a trained model with a detailed PDF report.

• Agent frameworks
• Models

A free, open-source anti-detect browser built as its own Chromium fork that spoofs its fingerprint in C++ rather than through injected JavaScript, with per-profile personas, proxies and an optional self-hosted team server.

• Browser automation

A web-based management dashboard for Weaviate and Qdrant vector databases, covering collection management, object CRUD, vector and text search, and cluster statistics.

• Databases

An open-source, local-first personal AI that remembers a user's goals and context across sessions and channels, and takes action on connected tools with confirmation required before high-impact steps.

• Agent frameworks

A dependency-free MCP and local tool gateway that curates and forwards tools from multiple MCP servers into one leaner interface, gates every call through a user's own policy hooks, and can package new MCP servers with no code.

• MCP servers

A set of agent skills that teach AI coding agents to generate and animate pixel art and HD game assets, build UI kits, and export finished animations into Godot projects using SpriteCook's MCP tools.

• Agent skills
• Generative media

A Convex component for semantic search over text chunks and embeddings, used to give an AI agent retrieval-augmented context, with per-user namespaces, custom filtering and importance weighting.

• RAG
• Databases

An MCP server that gives AI assistants semantic search over a local Qdrant vector database, with AST-aware codebase indexing, git history search, hybrid keyword and vector search, and support for several embedding providers.

• MCP servers
• RAG