Skip to content

cdif_maker.md

Latest commit

 

History

History
32 lines (22 loc) · 1.37 KB

cdif_maker.md

File metadata and controls

32 lines (22 loc) · 1.37 KB
Raw

🏷️ CDIF Maker Skill

The CDIF Maker skill enables the Croissant Toolkit to interact with specialized CDIF (Cross-Domain Interoperability Framework) semantic services. It allows for the discovery and structured mapping of variables into standard formats.

🌟 Key Features

  • Semantic Discovery: Translates natural language queries (e.g., "sea surface temperature") into precise scientific variables.
  • Variable Inventory: Generates structured variable inventories including definitions, units, and dimensions (time, depth, location).
  • Integration: Seamlessly outputs JSON files that can be consumed by the Croissant Expert for metadata enrichment.
  • Ollama Powered: Uses a dedicated Ollama model hosted at CODATA for high-precision semantic mapping.

🛠️ Usage

To discover variables and generate an inventory:

python3 .gemini/skills/cdif-maker/scripts/cdif_maker.py "variable name or description"

Example

python3 .gemini/skills/cdif-maker/scripts/cdif_maker.py "soil moisture at 10cm depth"

⚙️ Configuration

The skill connects to the CODATA CDIF service:

  • Endpoint: https://cdif-4-xas.dev.codata.org/ollama
  • Default Model: gpt-oss:latest

📂 Output

The results are saved as structured JSON files in the ./data/cdif/ directory, ready for integration into larger dataset metadata files.