MCP Tools Directory
← Browse integrations

Developer tools

io.github.neurarch-ai/neurarch-mcp

Free account · sign-in required

Contact io.github.neurarch-ai/neurarch-mcp

Loading secure sign-in…

Tools for autonomous AI agents

About this integration

Reads a PyTorch .py, a Neurarch graph, or a HF repo; agents inspect, lint, verify and rank designs.

Transport
stdio
Authentication
none
Initial setup
none
Runtime
unattended
Evidence
documented
Version
0.14.0
Package
neurarch-mcp
Last compatibility test
Not independently tested

Connect your agent

neurarch-mcp

Publisher documentation reviewed; package and model files not executed. Most tools run locally without credentials; optional plan/history service calls have separate network and key behavior.

Capabilities: Parse neural-network graphs, Lint architecture designs, Estimate readiness and deployment cost

Connected profiles

Additional details

Registry identifier

io.github.neurarch-ai/neurarch-mcp

Source ↗ · Checked 2026-09-17
Published version

0.14.0

Source ↗ · Checked 2026-09-17
Metadata license

CC0-1.0; package licenses are separate

Source ↗ · Checked 2026-09-17
Package ecosystem

npm

Source ↗ · Checked 2026-09-17
Source artifact

neurarch-mcp

Source ↗ · Checked 2026-09-19
Source manifest version

0.16.0

Source ↗ · Checked 2026-09-19
Unattended configuration

Your coding agent reads your model as source text, so it guesses at shapes, parameter counts, and what an edit breaks. `neurarch-mcp` hands it the structured graph instead, plus the verifier: **the design rules, the full readiness / cost / deployment verdict, and which of several candidate designs is worth the GPU time.** Trace the model to get real shapes, or point it at a saved graph, a PyTorch `.py`, or a Hugging Face repo. Every tool but two runs on your machine, offline, with no API key and

Source ↗ · Checked 2026-09-19
Authentication and prerequisites

Your coding agent reads your model as source text, so it guesses at shapes, parameter counts, and what an edit breaks. `neurarch-mcp` hands it the structured graph instead, plus the verifier: **the design rules, the full readiness / cost / deployment verdict, and which of several candidate designs is worth the GPU time.** Trace the model to get real shapes, or point it at a saved graph, a PyTorch `.py`, or a Hugging Face repo. Every tool but two runs on your machine, offline, with no API key and

Source ↗ · Checked 2026-09-19
Connection configuration

claude mcp add neurarch -- npx -y neurarch-mcp /abs/path/to/model.py

Source ↗ · Checked 2026-09-19
Review scope

Publisher documentation reviewed; package and integration endpoint not executed or independently security-audited.

Source ↗ · Checked 2026-09-19