A compact integration that uses LiteLLM to deliver flexible text completions via OpenAI models with configurable prompts and routing.

Secure HTTP-backed MCP server for Factorio, enabling Claude to run Lua, manage players, monitor, and capture screenshots via RCON-wrapped commands.

Rust async client for the Model Context Protocol enabling capability negotiation, tool execution, and resource management with Tokio-based async support.

TypeScript tool to manage Claude prompts via CLI, featuring creation, listing, and experimental argument support with desktop integration.

Lightweight Go-based SQLite example demonstrating protocol usage with simple config and quickstart tooling.

A lightweight CLI bridge exposing notes as resources via a custom cli:// URI, with add, help, and run tooling and an Inspector debugging workflow.

Modular MCP layer for Moqui ERP enabling private AI-assisted WeChat interactions with RAG-backed knowledge retrieval

Bedrock-based chatbot leveraging Claude 3.7 Sonnet, real-time MCP services, document processing, streaming replies, and cross-service orchestration for expert Q&A.

Securely run existing configurations without edits, with built-in secret vault and cross-platform launcher.

AI-driven checkout with OTP, wallet binding, merchant discovery, and 30-minute one-time virtual cards for secure online payments.

Give your AI a perfect, infinite memory. A local-first, zero-LLM memory system and Model Context Protocol (MCP) server designed to give AI assistants (like Claude, ChatGPT, and custom agents) a searchable, structured "Memory Palace."

A robust Model Context Protocol (MCP) server for scraping 1688.com products, bypassing anti-bot mechanisms using headed DrissionPage.

Provider-agnostic MCP server delivering private document search with flexible backends, local embeddings, and agentic results.

Bridges AI agents with SafeBreach to query simulators, test histories, and playbooks across multiple environments with secure access.

Telegraf-based bridge enabling LLMs to send, forward, and manage Telegram messages with intelligent sampling and access control.

Harvests React components from Aceternity and Shadcn, storing full code or metadata and generating ready-to-use integration prompts.

Lightweight, verifiable model context commitments with onchain memory, auditable registries, and gas-efficient metadata binding for AI systems.

Hybrid Aha MCP engine delivering offline-ready data with semantic search and flexible transport options.

Persistent concept map for codebases that lets LLMs reuse context across sessions, reducing token usage and speeding architectural reasoning.

Provides Claude direct access to a Raspberry Pi via a Cloudflare Tunnel with OAuth-based authentication.

Rust-based toolchain converts noisy HTML into clean, budget-conscious text with multi-backend search and optional LLM integration.

AI-driven engineering framework enforcing architecture-led planning, evidence-based validation, and deployment readiness.

AI-assisted module design with autonomous adventure creation and robust NWN tooling.

Self-hosted semantic memory with local embeddings in PostgreSQL, accessible as a Claude Code plugin via eight atelier tools.