Explore the directory

A command-line tool that connects to multiple robots at once, gives each one a large language model brain and a smaller decision model, and lets a user state a goal from a terminal or a chat to one robot or a coordinated group, called a flock.

• Agent frameworks
• CLI tools

A terminal UI that runs several AI agent sessions in parallel from one panel, each remembering its own context across turns, checking for output conflicts before acting, and asking for approval before risky actions.

• Coding agents
• Terminals

A fully offline private AI chat app for Android that runs any GGUF model on-device through llama.cpp with no network permission at all, plus optional Vulkan GPU acceleration and per-device optimized CPU kernels.

• Inference
• Mobile

A terminal UI for managing agent skills across a project, showing every installed skill in one place, which coding agents such as Claude Code, OpenCode or Cursor can actually see it, and why a skill's visibility might be broken.

• Agent skills
• Terminals

A command-line semantic grep for Apple Silicon that reranks or searches files by meaning using Jina embedding models running locally through MLX, with no PyTorch or cloud dependency.

• RAG
• CLI tools

A Chinese-language, source-code-based tutorial that walks developers through the DeepSeek Harness (dsh) agent framework chapter by chapter, from basic concepts to writing your own plugin.

• Coding agents
• Tutorials

An extension for the Pi coding agent that rewrites bash commands to their faster rtk equivalents and compacts noisy tool output from commands such as bash, read and grep, to reduce how much context window a session consumes.

• Agent skills

An early-stage operating layer for one-person businesses that coordinates a roster of specialized agents, such as scouts, engineers, designers, marketers and finance, as a single deliberating hive rather than independent agents taking turns.

• Agent frameworks

A Kubernetes operator that runs Ollama-served large language models on a cluster, letting a user deploy and manage multiple models the way they would manage any other Kubernetes workload.

• Inference
• Containers

A Flutter mobile app for iOS and Android that provides one chat interface for on-device LLMs and for local or cloud LLM servers such as LM Studio, Ollama and OpenRouter, with MCP tool support and a built-in model manager.

• Inference
• Mobile

A curated set of skills, provider profiles and configuration templates for OpenAI's Codex CLI, adding deep research, image generation, browser debugging through Chrome DevTools MCP, and quick switching between model providers such as Azure OpenAI, OpenRouter and GitHub Copilot.

• Agent skills

An MCP integration that connects QGIS to any AI agent speaking the Model Context Protocol, exposing 125 tools for managing layers, editing features, running processing algorithms, rendering and styling maps, authoring layouts and atlases, and running cross-layer SQL.

• MCP servers

A local-first agentic RAG system that turns PDFs, scans, DOCX, PPTX and XLSX into a searchable team knowledge base with enterprise access control, exposed through both an HTTP question-answering API and an MCP interface an agent can reason over.

• MCP servers
• RAG

A visible team workspace where multiple AI coding backends, such as Claude, Codex and Gemini, collaborate under 23 defined roles on one project, with a leader coordinating planning, coding, review and release across isolated Git worktrees.

• Coding agents

A local-first desktop workspace that gives each AI teammate, backed by Codex, Claude Code, Grok, OpenCode or Gemini, its own persistent workspace, task queue and conversation, with agent-to-agent messaging, file transfers and a shared embedded browser they can control.

• Coding agents

A command-line tool that turns a plain-English objective into a real Chrome browser automation run and returns structured results, used both as a standalone browser automation CLI and as a validation layer AI coding agents call to check their own work in a real browser.

• Testing
• Browser automation

A code-generation and package-management toolkit for embedded robotics projects built on the LibXR hardware abstraction layer, generating a C++ entry point from module configuration.

• CLI tools

A real-time instrument that visualizes what a language model is internally 'thinking' word by word during a live conversation, using the Jacobian lens on residual-stream activations.

• AI evals

An open-source, terminal-based AI coding assistant with agentic file search, editing and command execution, built as a Rust backend paired with a React and Ink terminal interface, supporting both cloud model APIs and local models through Ollama.

• Coding agents

A toolkit for training and evaluating German-language word embeddings, covering obtaining and preprocessing German text corpora, training models with gensim's word2vec, and evaluating them against generated test sets.

• Models

A native desktop app that gives a unified control room for Claude Code and Codex sessions: browse and search session history across projects, drive several parallel sessions from one workbench, and race the same prompt across different engines, models or channels side by side.

• Coding agents

An MCP server that exposes Milvus vector database functionality to LLM applications and AI coding tools such as Claude Desktop and Cursor.

• MCP servers
• Databases

A set of ten collaborating AI design agents that run a full, user-directed design process from discovery through shipping, checking accessibility at every phase rather than as a final step, with the user able to approve, correct or redirect at each handoff.

• Agent frameworks
• Design tools

A local-first memory layer that captures conversations from browser AI chats such as Claude or ChatGPT and from MCP-connected coding tools such as Cursor, builds a searchable local knowledge graph, and injects relevant context into new prompts either way.

• Agent memory