MCP server — Compressed Knowledge Graph for the NVIDIA AI developer stack. 20 domains · 998 nodes · 4× F1 vs RAG · 11× fewer tokens · deterministic traversal.
View on GitHub ↗MCP server — Compressed Knowledge Graph (CKG) for the full NVIDIA AI developer stack.
4× F1 · 11× fewer tokens · 998 nodes · deterministic traversal.
Read-only. This MCP server never writes, mutates, or executes. Every response is a declared graph traversal — not inference, not retrieval, not generation.
More agents. More retrieval. More context. And accuracy drops.
This is the intelligence paradox: the more AI you add, the more tokens you burn re-discovering structure your system already knows — or could know. Research finds 73% of enterprise tokens are redundant context. In multi-agent pipelines, context efficiency collapses from 18.2 in Q1 to 1.6 by Q4 — 91% degradation with no model change.
The model is not the bottleneck. The context is.
Every time your agent calls out to ask what TensorRT-LLM requires to run on Hopper, it spends ~2,982 tokens re-inferring a relationship that could be declared once and traversed in 269. That difference compounds across every query, every agent, every boundary crossing.
The fix is not a better model. It is structured context.
A CKG is a layer — a fast, inexpensive way to convert a large volume of domain documentation into structured, agent-traversable knowledge. Instead of retrieval, the agent traverses. Instead of inference, it reads declared relationships.
This package gives your agent the NVIDIA AI developer stack as its first layer: 20 domains, 998 nodes, every prerequisite chain declared and typed.
From the project README.
Add the radar badge to your README — it shows your project was picked up by MCP Radar and links to this page:
[](https://mcp.liqiwa.com/s/Yarmoluk--ckg-nvidia-ai.html)
Independent, source-backed handbook for DeepSeek AI's official DeepSeek Harness (dsh): agents, plugins, security, troubleshooting, and runbooks.
shitianfang/jev-useClaude Code / Codex / pi plugin that hands agent steps needing no text output to Jev (TypeSafe's judgment model) — measured p50 ~230 ms and ~$0.02 per 1,000 judgments, with typed escalation back to th
abluva/mcp-remoteThe bridge for remote MCP — seamless OAuth, resilient auth recovery, production-grade reliability
graphersal/graphersalIn-memory graph database and Gremlin traversal engine in Rust: browser playground, CLI, MCP server for AI agents, durable store, Python bindings.
shinegang/baronBaron Munchausen — local memory for coding agents that outlives the chat, with a verdict on every answer. Public alpha.
replenhq/replenSmarter AI Development Workflows
The top new MCP servers of the week, every Monday. No spam, unsubscribe anytime.