MCP Servers

Cybernetic control plane for agentic systems with governed context, memory, safety, and execution

CLI and MCP tools for Yorck Berlin cinema.

Drop-in screenshot gateway for UI agents. Filters redundant frames, extracts semantic state changes, and cuts vision-token spend. Ships as an MCP server, an in-process routing library, and a local Anthropic-compatible proxy. Measured 70-82% input-token an

SQLite-backed durable task queue with first-class MCP integration, purpose-built for AI coding agents (crash recovery, audit trail, embedded daemon).

MCP server for the Waikay brand visibility API

CLI to initialize and manage Cortex Context (Knowledge Graph) in any VS Code workspace

Local-first Kubernetes-style runtime for autonomous coding agents

MCP server for Business Central AL performance pattern analysis — detects 38 anti-patterns with SQL impact data and workshop hints

Shared primitives for the AI-agent governance suite: Finding schema, JSONC/TOML readers, line locators, MCP command normalization, shell tokenization, transcript-event types + JSONL parsers (Claude Code/Cursor/Codex), and GitHub Action helpers.

Open-source CLI to minimize LLM token usage via smart code context extraction

The trustworthy FHIR bridge for AI agents — MCP server with reproducible clinical evals, reversible de-identification, and CDS Hooks.

MCP server for lovable.dev — install with zero credentials; Claude collects them at first use via lovable_login. Drives the dashboard, AI agent, linked GitHub repo, Supabase, and deployed-env secrets. ~58 tools, stdio transport.

Model Context Protocol server for Venice.ai — uncensored, private, crypto-native AI inference for Claude, ChatGPT, Cursor and any MCP host. Community-maintained. Provided as-is, with no warranty or SLA from Venice AI. Use at your own risk.