What is local-llm-mcp-server CLAUDE.md?

A set of project instructions for an AI coding agent working in a monorepo—a single repository containing several related projects. It describes a production MCP server and six machine-learning research projects for NVIDIA DGX Spark computers.

Use it when asking an agent to understand, modify, or deploy the MCP server or its machine-learning subprojects. It is also a reference for the projects exploring neural networks, continuous-time models, and ARC-AGI, a benchmark for solving visual reasoning tasks.

It gives the agent the project structure and relationships it needs to work in the right part of the repository. It also documents deployment details and how the research projects fit together.

How georgepok/local-llm-mcp-server configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to georgepok/local-llm-mcp-server. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/georgepok/local-llm-mcp-server/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/georgepok/local-llm-mcp-server

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/georgepok/local-llm-mcp-server/claude-md/github.svg)](https://agentmods.dev/instructions/georgepok/local-llm-mcp-server/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/georgepok/local-llm-mcp-server/claude-md"><img src="https://agentmods.dev/badge/instructions/georgepok/local-llm-mcp-server/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/georgepok/local-llm-mcp-server/claude-md"><img src="https://agentmods.dev/badge/instructions/georgepok/local-llm-mcp-server/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 1,658 This file is loaded in full into every session.

When invoked 1,658 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.01658	$0.01658
Opus 5.5	$0.00663	$0.00663
Sonnet 5.5	$0.00332	$0.00332
Haiku 4.5	$0.00166	$0.00166

Measured 2mo ago against content hash 91ae9e5c76b5, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

local-llm-mcp-server CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 145 lines

How it starts

The opening of the file, as written. The whole thing — 145 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Code Project Guide

What This Is

A monorepo combining a production MCP server with six research subprojects exploring geometric neural networks, continuous-time ODE models, and ARC-AGI solving — all targeting NVIDIA DGX Spark (GB10, 128GB unified memory).

Repository Structure

local-llm-mcp-server/
├── src/                    # Production MCP server (TypeScript/Node.js)
├── config.json             # MCP server config (points to DGX Spark vLLM)
├── subprojects/
│   ├── fluid-geometry/     # vLLM logits processor — entropy-driven reasoning control
│   ├── fgn-v3/             # Fluid Geometry Network — Riemannian attention replacement
│   ├── liquid-arc/         # LiquidARC — continuous-time geometric ODE for ARC-AGI
│   ├── latent-oracle/      # Oracle embedding distillation via HyperNet
│   ├── neuroplastic/       # Self-introspecting neural substrate (Nemotron)
│   └── wake-sleep/         # Autonomous wake-sleep training loop

Subproject Relationships

The subprojects form a research progression:

fluid-geometry — Production plugin. Single-file vLLM LogitsProcessor that adapts temperature and thinking tokens based on entropy curvature. Deployed on DGX Spark.
fgn-v3 — Foundation research. Replaces transformer dot-product attention with diffusion on learned Riemannian manifold. Heat kernel attention via geodesic distances. This is the theoretical basis for LiquidARC.
liquid-arc — Primary research project. Clean-slate continuous-time ODE model where geometry IS the computation. One shared dynamics module applied 16x via Euler integration. SDPA-factored heat kernel enables FlashAttention. Key results:
- Phase transition at CV~6.0 produces universal geometric substrate
- Geometry distillation: 71% eval ARC (vs teacher's 54%), 72% graph coloring
- 100x geometric LR ratio is the critical training insight
latent-oracle — Distillation. Uses oracle model (Qwen3.5-9B) embeddings as task descriptors to predict task-specific weight deltas via HyperNet.

Read the full file on GitHub · 145 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 145 lines · 1,658 tokens per session scan A 91ae9e5c76b5

Subscribe to this mod's changes

local-llm-mcp-server CLAUDE.md is an instructions file published in the GitHub repository georgepok/local-llm-mcp-server (0 stars, last pushed 3mo ago), licensed MIT. It adds 1,658 tokens to every session, about $0.0066 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-01.

local-llm-mcp-server CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
local-llm-mcp-server CLAUDE.mdgeorgepok/local-llm-mcp-server	Instructions file	0	3mo ago	MIT	A	1,658	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

local-llm-mcp-server: Instructions file for Claude Code