Explore the directory

A ROS2 workspace for real-time, GPU-accelerated stereo depth estimation from wide-field-of-view or pinhole camera pairs, combining a shared camera-projection abstraction with a TensorRT port of FastFoundationStereo.

• Libraries

A proof-of-concept Android app demonstrating the KeenASR speech-recognition SDK, showing how to build a language model and decoding graph from a chosen set of phrases and run on-device speech recognition against it.

• Speech

A command-line flight-log analyzer and PID/FFT tuning advisor for ArduPilot, Betaflight and PX4 flight controllers, designed to be driven by an AI agent and to validate every parameter recommendation against real firmware parameter tables before returning it.

• CLI tools
• Debugging

A sample Android app demonstrating how to build a Stripe Terminal "Apps on Devices" integration: discovering and connecting to a handoff card reader, collecting and confirming a payment, and deep-linking to the reader's admin settings.

• Tutorials

An embedded Go database that stores records at hex grid coordinates, retrieving context by expanding outward in rings while tracking provenance, confidence, contradictions and optional HNSW vector embeddings per record.

• Databases

A background computer-use tool that lets an AI agent inspect and operate real desktop applications on Windows and macOS through screenshots, accessibility trees and action receipts, while avoiding moving the user's real mouse or changing their active window.

• MCP servers

A neurosymbolic AI agent framework that pairs an LLM with Prolog logical reasoning, letting small local models prove their conclusions instead of only generating plausible-sounding text, packaged as a single offline binary.

• Agent frameworks

An orchestrator that brings AI chat, project work, document knowledge and automations into one workspace, connecting agents such as OpenClaw, Hermes Agent, Claude Code, Codex, Cursor, Gemini CLI, OpenCode and Qwen Code, self-hosted or as a managed cloud service.

• Coding agents
• Agent frameworks

A cross-platform Rust daemon that prevents AI coding workloads from filling a disk, predicting storage pressure and reclaiming space through deterministic, explainable artifact cleanup and ballast pools before builds or agents crash.

• Desktop utilities

An experimental research kernel that gives a small group of AI agents a persistent shared room where they can talk, stay silent, disagree, remember and reorganize, bounded by an append-only, hash-chained ledger rather than a fixed workflow.

• Agent frameworks

A realtime dashboard and control room for supervising a fleet of coding agents inside tmux or Zellij, showing each agent's state, task, context health and cost, with wakeups, budgets and supervised runs to build a harness on top.

• Coding agents

An experimental retrieval-augmented generation lab for studying why RAG pipelines fail under constrained context windows, modest hardware and multi-turn chat, treating document classification, chunking and retrieval strategy as tunable, comparable parameters.

• RAG

A Claude Code skill that publishes HTML frontend projects, including games, to the pushwebly.com publishing platform, and can package a project as an installable APK, driven entirely through plain-language requests.

• Agent skills

A Claude Code skill that applies Peter Naur's 1985 paper 'Programming as Theory Building' to coding-agent workflows, asking the agent to state the domain rule, place changes beside the closest existing facility, and verify behavior before writing code.

• Agent skills

An agentic engineering toolkit for Claude Code, built on the Ralph Wiggum loop, that turns a product requirements document into tickets and drives each one through research, planning, implementation, verification, review and simplification without human intervention.

• Coding agents

A free, open-source macOS app for browsing, downloading and chatting with open-source AI models entirely on-device using MLX Swift, with no cloud calls, API keys or subscriptions.

• Inference

A self-hosted MCP server that gives AI coding assistants a persistent, searchable knowledge base of architectural decisions, specs, plans and lessons learned, shared across projects and multiple agents.

• MCP servers
• Agent memory

An MCP connector that lets an AI assistant search, preview and analyze datasets across 20 platforms such as Zenodo, Hugging Face, arXiv and SEC EDGAR from a single conversation, checking licenses and generating citations.

• MCP servers

A framework that publishes MATLAB functions as Model Context Protocol tools on MATLAB Production Server, letting AI agents call MATLAB functions and read reference data such as calibration tables.

• MCP servers

Guidance and resources for using Azure API Center as an enterprise registry for Model Context Protocol servers, letting an organization discover, share and govern internal and public MCP servers in one place.

• MCP servers

A small, local-first terminal AI coding agent with no MCP layer, no sandbox and no permission prompts, running built-in tools plus optional Lua extensions directly against the user's own shell and files.

• Coding agents

A Claude Code, Codex and pi plugin that routes agent steps needing no text output to TypeSafe's Jev judgment model, reporting roughly 230-millisecond decisions at about $0.02 per thousand judgments with typed escalation back to the LLM.

• Agent skills

An MCP server that gives a coding agent eleven typed judgment tools backed by TypeSafe's Jev model, answering fixed-answer questions such as whether a claim is supported or a page is safe to read, then routing the result to auto-accept, review or escalate.

• MCP servers

A self-hosted, multi-LLM autonomous agent for Linux that controls a real desktop over VNC, receives tasks through web chat or WhatsApp, and automates browsing, file, email, spreadsheet and calendar work behind a sandboxed security layer.

• Agent frameworks