Explore the directory

A set of deliberately vulnerable machines, grouped by the skill each one exercises, for practicing web application security testing.

• Security

A pure-Go, drop-in Redis-protocol-compatible in-memory data store with native HNSW vector search, Redis Streams, cluster gossip, and embedded Lua and WebAssembly scripting runtimes.

• Databases

A high-performance C++ multi-model database combining graph, vector, relational and file storage under one query language with MVCC, aimed at applications that need native AI and LLM data integration.

• Databases

A memory-compression system for long-context transformer inference that combines finite scalar quantization with dynamic latent-space mapping to shrink key-value cache tensors, with an open orchestration layer and a separately licensed core compression binary.

• Inference

A local sidecar that turns an AI agent's completed execution traces into reviewed memories, reusable skills and evidence-backed training data, gating every proposed change behind human review instead of silently mutating agent state.

• Agent memory
• AI evals

A full-stack retrieval-augmented-generation template covering authentication, knowledge-base management, document parsing and vectorization, hybrid retrieval, streaming chat and evaluation in one repository.

• RAG

A locally deployable, open-source multimodal mixture-of-experts language model focused on financial applications, compressed to under 16 GB through a custom quantization method while running on consumer GPUs.

• Models

A C++ and Metal extension for MLX on Apple Silicon that adds native K-quant quantization, alongside a toolchain to quantize, load, run, LoRA-train and fuse K-quant checkpoints on a stock MLX installation.

• Inference

A self-updating index of Model Context Protocol servers published on GitHub, refreshed every 15 minutes by a GitHub Actions cron job that ranks them by real activity such as popularity, commits and release freshness.

• Curated lists

A local-first desktop dashboard for Windows and macOS that runs Claude Code, Codex and a built-in ReAct engine in one window, with shared long-term memory, MCP tools and no account or relay server.

• Coding agents

A single-binary, zero-dependency AI coding assistant for the terminal, written in Go, supporting Anthropic, OpenAI and Ollama backends with built-in tools, subagents, an approval gate and automatic post-edit build verification.

• Coding agents

A community-built serving stack that runs NVIDIA's GLM-5.3-Flash NVFP4 checkpoint across two DGX Spark-class GB10 systems through a pinned vLLM build, aimed at commercially usable, pinned and reversible deployment rather than peak throughput.

• Inference

A native desktop app for macOS and Windows that gives ChatGPT controlled access to a local workspace through MCP, reading and modifying files, running git operations and, when explicitly enabled, local shell commands.

• MCP servers

A single static binary that discovers every skill, MCP server, plugin, hook and rule installed across a machine's AI coding tools, hashes them into a lockfile, statically scans them, and detects post-audit modification before it can cause harm.

• Security

A demonstration Model Context Protocol server that exposes ERP-style sales, inventory and purchasing data to AI agents, with a genuinely enforced human-in-the-loop approval gate on write actions and a full audit trail.

• MCP servers

A private AI assistant and local runtime that runs entirely on the user's own hardware: it holds conversations by voice or text, controls the desktop, reads the screen, writes code, and builds a private model of the user over time.

• Agent frameworks
• Agent memory
• Speech

An OAuth 2.1 Streamable HTTP MCP client plugin for DeepSeek Harness that adds PKCE, dynamic client registration, browser-based login and persistent token storage to the native MCP client connection flow.

• MCP servers

A DeepSeek Harness plugin that replaces loading every connected MCP server's full tool schema with two model-facing tools, one that searches across servers for relevant tools and one that calls a specific tool by name.

• MCP servers

An MCP server that remote-controls a full RAD Studio Delphi installation, its language server, build system and deploy chain, so an AI agent can develop, build and deploy Delphi projects from any platform, including driving Android, Linux and Windows targets.

• MCP servers

An independent, reproducible benchmark that compares APIs choosing one option from a list and returning a confidence score, measuring accuracy, latency, cost, failure modes and calibration across a decision model and several constrained LLMs.

• AI evals

A spec-driven development framework of commands, scripts and skills for AI coding assistants that turns a feature idea into discovery, planning, review and delivery through a fixed sequence of slash commands.

• Agent skills

An autonomous penetration-testing MCP server with an integrated AI decision engine, designed to run on a Raspberry Pi running Kali Linux ARM64 and chain pentest tools with minimal LLM calls.

• Security

A structured, decision-grade evidence base on long-term memory for LLM agents, combining a 989-paper index, archived source pages, product notes, benchmark protocols and architecture maps.

• Papers

A self-hosted multi-agent platform that audits a website's technology stack, then provisions a fleet of specialist AI agents to plan, execute and verify work against that infrastructure, with automatic model-provider failover.

• Agent frameworks