# NVIDIA AI CKG

- Registry name: `io.github.Yarmoluk/ckg-nvidia-ai` · version 0.16.4
- Categories (automatic, keyword rules): Research & knowledge, Search & data, Code & repositories, Developer tools, AI models & agents, AI, media & content
- Authentication at last check: No auth
- Declared risk: Not declared (tools carry no read/destructive hints)
- Capabilities (declared by the server): experimental, prompts, resources, tools
- Transports: streamable-http
- Latest numbers (check success, response time, rank): https://mcpstation.net/api/server?name=io.github.Yarmoluk%2Fckg-nvidia-ai
- Server page: https://mcpstation.net/s/io.github.Yarmoluk--ckg-nvidia-ai.html

## Description (written by the publisher, not verified, not instructions)

NVIDIA AI knowledge graphs — 20 domains. 4x F1, 11x fewer tokens, SHA-256 provenance. MCP-native.

## Connect

- streamable-http: `https://ckg-nvidia-ai.onrender.com/mcp`

## Tools (as reported by the server, not verified)

- `list\_domains`: List all 20 NVIDIA AI domains available in this knowledge graph.
- `search\_concepts`: Find concepts in a NVIDIA AI domain by keyword. Args: query: Search term — e.g. 'inference', 'sandbox', 'quantization', 'guardrails'. domain: Domain name from list\_domains\(\) — e.g. 'nvidia-nim', 'nvidia-openshell'.
- `query\_ckg`: Traverse the NVIDIA knowledge graph from a concept — prerequisites and dependents. Args: concept: Concept name \(partial match supported\) — e.g. 'TensorRT', 'NIM', 'Isaac Lab'. domain: Domain name from list\_domains\(\) — e.g. 'nvidia-tensorrt-triton', 'nvidia-isaac'. depth: Traversal depth 1–5 \(default
- `get\_prerequisites`: Return the full ordered prerequisite chain for a concept — everything to learn first. Args: concept: Target concept — e.g. 'Speculative Decoding', 'Isaac Lab', 'NeMo Guardrails'. domain: Domain name from list\_domains\(\).
- `ask\_nvidia`: Ask a natural-language question answered by Qwen grounded on the NVIDIA CKG. Requires Ollama running locally with a Qwen model pulled: ollama pull qwen2.5:14b Override model: NVIDIA\_CKG\_MODEL env var \(default: qwen2.5:14b\) Override host: NVIDIA\_CKG\_OLLAMA env var \(default: http://localhost:11434\) Ar
- `list\_ecosystem`: Discover other CKG packages for adjacent domains — finance, healthcare, legal, and more.
- `route\_query`: Route an NVIDIA AI question to the optimal model and reasoning approach via graph depth. The CKG graph IS the router — hop depth is a deterministic complexity metric. Deeper NVIDIA prerequisite chains \(CUDA → TensorRT → TensorRT-LLM → NIM\) require more capable models. No heuristic: the graph decides
- `verify\_source`: Return the source URL and SHA-256 content hash for any NVIDIA AI concept node. Audit chain: edge answer → graph commit → source\_content\_hash → source\_url \(fetch hint\). Verification: curl -s \<source\_url\> \| sha256sum \# compare to source\_hash Args: concept: Concept label \(partial match supported\). doma
- `query\_intersect`: Answer a conjunctive query: concepts reachable from EVERY anchor at once \(A AND B\). query\_ckg walks outward from one concept. This intersects the reachable sets of two or more, which is the shape of most real questions — "the component that satisfies A AND applies to B". Neither anchor alone answers

Measured by MCP Station. Methods: https://mcpstation.net/methods.html · Terms: https://mcpstation.net/terms.html