31 tagged token-efficiency, measured the same way as everything else here.
Browse within: Structured Output 26ai-coding 26mcp-server 26
MCP server Claude CodeCodexCursor +2
One of 26 in server.json
Structured git operations (status, log, diff, branch, show) as typed JSON. Runs locally from the @paretools/git npm package.
Fractera/Agent-Engineering-Infrastructure
MCP server Claude CodeCodexCursor +2
MCP server "fractera", hosted remotely at www.fractera.ai, as configured in Fractera/Agent-Engineering-Infrastructure.
MCP server Claude CodeCodexCursor +2
Vendor-neutral MESI + optimistic-concurrency coordinator for multi-agent state — a stale write is denied or returned as a typed conflict, never silently applied; safety invariants model-checked in TLA+. Runs locally from the agent-coherence Python package.
MCP server Claude CodeCodexCursor +2
Drop-in MCP proxy. 71% fewer tokens. Session dedup compounds to 92%. Zero code changes. Runs locally from the @blackwell-systems/gcf-proxy npm package.
MCP server Claude CodeCodexCursor +2
MCP server "toke-local" as configured in rianprei/toke-local. Runs locally from the toke-local npm package.
MCP server Claude CodeCodexCursor +2
A token-disciplined developer MCP hub with a tiny always-loaded surface and deferred internal capabilities. Runs locally from the tokenhub-mcp npm package.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: