Explore the directory

A persistent memory system for AI agents that continuously transforms raw conversations into a knowledge graph of atomic facts and their relationships, consolidating and superseding stale facts in the background rather than only retrieving similar text chunks.

• Agent memory

An MCP server and command-line tool that makes a folder of Markdown notes searchable by meaning rather than only by keyword, splitting notes into chunks, embedding them with a locally or remotely configured model, and indexing the result in a local FAISS store.

• MCP servers
• RAG

An open-source JupyterLab extension and desktop app that brings a Cursor-style AI editing experience directly into notebook cells, with inline edits, a multi-step agent that can read, edit and run cells, ghost-text completion, and one-click traceback fixes.

• Editors

An agent harness that runs one user-defined agent, with its models, persona, tools and permissions set in a single JSON file, across many surfaces: a terminal REPL, a one-shot script command, a scheduled job, or a Telegram or Discord bot the user hosts themselves.

• Agent frameworks

A single-command terminal benchmark that discovers the language models already installed in a local runner, such as Ollama, LM Studio, llama.cpp or vLLM, and measures their speed, memory use and answer quality together on a live leaderboard.

• Inference
• AI evals

An enterprise, multi-tenant Agentic RAG knowledge-base system with a Vue.js frontend, a Spring Boot backend and a FastAPI LLM service, where the model chooses its own retrieval tools across several rounds and exposes knowledge bases to agents over MCP.

• RAG

A Python library for migrating a vector index to a new embedding model progressively, serving search from the existing index while target vectors are computed and cached in the background, instead of re-embedding the whole corpus before the new model can serve any traffic.

• RAG
• Databases

A self-hosted, asynchronous Telegram bot that orchestrates more than a dozen AI providers and their tools to carry out tasks such as code changes, research and media generation, running sub-agents in parallel and reporting back through the same chat interface.

• Agent frameworks

A local-first personal knowledge app for developers that imports past conversations from tools such as Claude Code, Cursor and Codex CLI, distills them into searchable notes with a tag knowledge graph, and exposes that knowledge to agents through MCP.

• MCP servers
• Agent memory

A four-hour, hands-on course that builds a complete AI nutrition-assistant agent with OpenAI's Agents SDK, covering tool calling, retrieval-augmented generation, the Model Context Protocol, multi-agent orchestration, memory, guardrails and cloud deployment.

• Tutorials

A Python framework for building and deploying production AI agents with a small amount of code, including type-safe structured outputs, built-in observability, automatic REST API generation, and Model Context Protocol integration.

• Agent frameworks

An embedded vector database for multi-agent memory that branches a base vector index in about half a millisecond using copy-on-write, instead of fully copying the index to snapshot, fork or checkpoint it as other vector stores require.

• Agent memory
• Databases

A fast, local, indexed code search tool written in Rust for Tantivy full-text indexing, built to be called by AI coding assistants, with a management terminal dashboard and a background indexing service.

• CLI tools

An agent harness for AI coding agents built around treating tool output as evidence and language-model output as an unverified claim, with execution sandboxing, seven interchangeable reasoning strategies, and a published, reproducible SWE-bench Lite score.

• Coding agents

A plug-and-play Docker image that runs vLLM with DFlash block-diffusion speculative decoding on NVIDIA Blackwell hardware such as the DGX Spark, optionally combined with TurboQuant KV-cache compression, to speed up decoding on Blackwell-exclusive NVFP4 tensor cores.

• Inference
• Containers

A local code-indexing MCP server for AI coding agents that combines SQLite full-text search, tree-sitter parsing and embeddings into a single searchable index, replacing many rounds of grep and file reads with direct symbol, call-graph and semantic queries.

• MCP servers
• RAG

An MCP server that gives Claude semantic search over an Obsidian vault by reusing the embeddings the Smart Connections Obsidian plugin already generated, running the same embedding model locally so a query never leaves the machine.

• MCP servers
• RAG

A plugin for the Hermes Agent framework, installed through Hermes's own plugin hooks without changing its core code, adding model-tier routing by task difficulty, a circuit breaker for repeated failures, and a self-healing error-pattern library.

• Agent frameworks
• LLM gateways

An open-source Python package that produces typed, calibrated decisions from any open-weights language model in a single forward pass, reading the model's own next-token probabilities rather than generating text, as an open alternative to TypeSafe's Jev.

• Inference
• Libraries

A self-hosted control plane for operating fleets of OpenClaw and Hermes AI agents on infrastructure the operator controls, deploying to Docker or Kubernetes with lifecycle management, live logs and a terminal, budgets and scheduled runs, and encrypted secrets.

• Containers
• Hosting

A Python library that runs structured natural-language tasks, such as extracting bullet points, generating question-and-answer pairs and building knowledge graphs, locally through a fine-tuned 135-million-parameter language model instead of a large general-purpose one.

• Models
• Libraries

The reference implementation of a research paper on collaborative self-evolution for LLM-based multi-agent systems, where a team of agents reflects together after each task at the level of individual agents, their interactions, and the team's own organization.

• Agent frameworks
• Papers

A zero-infrastructure, async-first Python library that gives AI agents persistent memory stored as plain Markdown files and indexed by SQLite, so every memory can be opened, searched, edited and tracked with git diff rather than living in an opaque database.

• Agent memory

A maintained fork of the mcp-remote stdio proxy that bridges MCP clients such as Claude Desktop to remote MCP servers over OAuth, fixing open upstream bugs around mid-session token expiry, callback hangs and localhost callback failures.

• MCP servers