MLOps

RunsNative MCP server — gives any MCP-capable agent access to RunsNative component docs, foundations, design exercises, and theme inference.

Give AI agents TypeScript type awareness

MCP server for MLoop CLI - ML.NET MLOps tool

Consciousness architecture plugin for OpenClaw implementing GWT, IIT, AST, and intrinsic motivation

Hugging Face MCP server for local inference and Hub repository management.

Database-Agnostic Backend Core for Rebase

TypeScript SDK for NocturnusAI — a logic-based inference engine and knowledge database

0G Code to Coin - MCP server for decentralized AI compute and storage on 0G Network

JSON Schema builder and validator for TypeScript with static type inference, Hono middleware for OpenAPI generation and validation, and MCP server/client implementation. Lightweight, dependency-free, and built on Web Standards.

curl for x402 paid APIs. Auto-pays any endpoint on Base and Solana.

Modal GPU Cloud MCP server - manage GPU workloads, training, and robot inference

Dual Memory Agent Runtime: Causal Graph Engine + General Memory + Forward Simulation + Backward Inference for AI Agents

Deterministic, zero-token CLI for Pipeline. Mirrors the hosted MCP server, but cron-safe, pipe-friendly, and exit-code-correct.

Browser-only knowledge management core: PGlite (PostgreSQL WASM) data layer, single-writer worker, MCP tools, job queue, capability system, SKOS taxonomy, hybrid search. 100% JSON format parity with the fortemi Rust server.

Tenzro Network SDK — wallets, identity, agents, inference, bridge, crypto, TEE, ZK proofs, custody, and settlement

Classify a GPU workload as inference or training from telemetry alone, no application changes required.