CODATA
Official@codata · Paris
CODATA is the Committee on Data of the International Science Council (ISC).
Agent Skills by CODATA
Showing 25 vetted skills indexed across 1 GitHub repositories.
creator
Generate visual presentations and MP4 videos from Markdown slides using Python.
transcriber
Download YouTube transcripts and retrieve video metadata with Python scripts.
Kreuzberg
Extracts text, metadata, and structured information from diverse file formats into Markdown with tables and code blocks.
obsidian_expert
Convert Croissant JSON-LD datasets into Markdown notes with embedded metadata.
rohub
Deposit research objects into RO-Hub with semantic annotations and provenance metadata.
telegram_expert
Send notifications and upload files to Telegram chats via the Bot API.
photograph
Capture web page screenshots and record screen activity using Playwright and ffmpeg.
nlp_expert
Extract named entities from text into standardized JSON-LD using Gemini 3 NLP.
wizard
Convert raw data sources into Croissant JSON-LD dataset metadata.
test-all
Discovers and runs all test_*.py scripts in skills directories, reporting successes and failures.
translator
Detect source languages and translate text or video scripts into English.
communication_officer
Send Croissant dataset files and summaries via SMTP with TLS.
presentation_expert
Convert research data into Markdown presentation slides with Python automation.
github
Connect to GitHub repositories to verify ownership and discover embedded skills.
architect
Create Mermaid.js system architecture diagrams from descriptive texts.
unf
Generate order-invariant data fingerprints for integrity verification using Python and Polars.
navigator
Automate Google searches and extract URLs and snippets into JSON files.
tester
Automate end-to-end testing of Croissant Toolkit skills and scripts.
croissant_expert
Generate and validate Croissant-compliant JSON-LD dataset metadata.
youtuber
Search YouTube by keywords and extract video titles, URLs, and descriptions.
ro-crate-expert
Automates creation of RO-Crate packages from Dataverse datasets via Python scripts.
login
Unlock protected development skills using stored authorization keys.
orchestrator_expert
Integrate and sequence multiple AI skills into automated data pipelines.
neo4j_expert
Convert Croissant dataset JSON-LD files into a Neo4j knowledge graph.
Frequently Asked Questions About CODATA
FAQPage SchemaWhat specific tasks can I perform with these research data management capabilities?▼
You can generate Croissant-compliant JSON-LD metadata, create RO-Crate packages from Dataverse repositories, build Neo4j knowledge graphs from structured datasets, and verify data integrity using order-invariant fingerprints.
Which technical personas benefit most from these research data management skills?▼
These capabilities are designed for data scientists, research software engineers, and information architects who require standardized metadata generation and semantic interoperability for scientific datasets and digital research objects.
What are the primary prerequisites for deploying these data management functions?▼
Deployment requires a local environment configured for execution of modular logic, access to target data sources like Dataverse or JSON-LD files, and valid authorization keys for secure repository and service integration.