MCP server that intercepts and compresses tool outputs before they reach the LLM context. Reduces 60-90% token consumption without losing
Free
1 servers · 1 listing
By vmexicano
Scanned where there is a package or repository to read, and every release diffed against the tool surface we already hold.
MCP server that intercepts and compresses tool outputs before they reach the LLM context. Reduces 60-90% token consumption without losing