"author:aaron-devx" matching MCP servers:

Enables AI coding assistants to compress tool schemas, file contents, JSON responses, and screenshots before they reach the LLM, with measured token savings, symbol-aware navigation, and SQLite-backed caching.