Datathings Marketplace
We are Datathings, specializing in high-performance software for large-scale data infrastructure. Our foundation is GreyCat — a temporal graph database and programming language built for efficiency at scale, with native agentic AI capabilities. On that foundation, we built Kopr: a digital twin managing Luxembourg's entire electricity distribution grid — 1 million grid assets, 330,000 delivery points, and 45 billion meter readings per year, with machine learning running continuously over live sensor data.
The plugins here bring that stack to your AI agent: GreyCat's runtime and language tools, the numerical and GPU computing libraries behind high-performance inference and optimization, and widely-used power systems analysis frameworks for anyone building in that domain.
Quick Start
Install GreyCat:
Linux, Mac or FreeBSD (x64, arm64):
curl -fsSL https://get.greycat.io/install.sh | bash -s dev
Windows (x64, arm64):
iwr https://get.greycat.io/install_dev.ps1 -useb | iex
⚠️ Important: Restart your terminal after installing GreyCat
Install Claude Code:
Follow the installation instructions at https://code.claude.com/docs/en/setup
Add the marketplace:
/plugin marketplace add datathings/marketplace
Install plugins:
/plugin install greycat@datathings
Try it:
Run Claude in a demo folder, then paste this prompt:
Use your greycat skill to create a GreyCat backend with Country, City, Street, House, and Person nodes linked as a geographic hierarchy with back references for bidirectional navigation (country contains cities, cities contain streets, etc., and children reference their parents). Add geo coordinates (latitude, longitude) to appropriate nodes. Houses should have temperature sensors storing time series data. Generate two sample CSV files: `./data/addresses.csv` (with house IDs) and `./data/temperatures.csv` (with house_id, date, value columns), and create an importer that loads both on startup (import the CSVs on main if the country index size is 0). Expose all important API endpoints. Create comprehensive API documentation and expose meaningful functions as MCP.
Plugins
| Plugin | Category | Type | Version | Description |
|---|
| GreyCat Technology | | | | |
| greycat | GreyCat Technology | Skill | 2.4.8 | Full-stack GreyCat development — GCL language, graph persistence, LSP, frontend integration |
| greycat-c | GreyCat Technology | Skill | 2.4.8 | GreyCat C API and Standard Library for native development |
| Agentic AI | | | | |
| llamacpp | Agentic AI | Skill | 2.4.8 | llama.cpp C API reference (163 functions) for local LLM inference |
| ggml | Agentic AI | Skill | 2.4.8 | ggml C tensor library (560+ functions) for graph computation, GGUF I/O, multi-backend inference, and ML training |
| vllm | Agentic AI | Skill | 2.4.8 | vLLM v0.16.0 — high-throughput Python LLM inference with offline batch, OpenAI-compatible server, LoRA adapters, multimodal inputs, and structured outputs |
| ollama | Agentic AI | Skill | 2.4.8 | Ollama v0.16.3 — run and interact with local LLMs via REST API (chat, generate, embed, model management) |
| High Performance Math & GPU | | | | |
| blas_lapack | High Performance Math & GPU | Skill | 2.4.8 | CBLAS & LAPACKE C API reference (1284 functions) for numerical linear algebra |
| cuda | High Performance Math & GPU | Skill | 2.4.8 | NVIDIA CUDA C/C++ — Runtime API, cuBLAS, cuFFT, cuSPARSE, cuRAND, cuSolver, Thrust, Cooperative Groups |
| opencl | High Performance Math & GPU | Skill | 2.4.8 | OpenCL SDK (Khronos) — cross-platform GPU/CPU parallel computing, C API (~60 functions), C++ wrapper (opencl.hpp), SDK utilities |
| rocm | High Performance Math & GPU | Skill | 2.4.8 | AMD ROCm 7.2.0 — HIP kernel development, rocBLAS/rocFFT/rocRAND/rocSOLVER libraries, profiling, and CUDA-to-HIP porting |
| Power Grid Management | | | | |
| pandapower | Power Grid Management | Skill | 2.4.8 | pandapower v3.4.0 — Python power systems analysis with AC/DC power flow, OPF, short circuit (IEC 60909), state estimation, and visualization |
| powergridmodel | Power Grid Management | Skill | 2.4.8 | power-grid-model v1.13.10 — high-performance Python library for steady-state distribution power system analysis: power flow, state estimation, and IEC 60909 short-circuit calculations |
GreyCat Technology
