Design, orchestrate, and execute LLM-powered data pipelines from YAML definitions.
npx skills add https://github.com/ucbepic/docetl --skill docetl
Please help me install this Agent Skill. Skill: docetl Source: https://github.com/ucbepic/docetl/tree/main/.claude/skills/docetl Command: npx skills add https://github.com/ucbepic/docetl --skill docetl
DocETL provides a framework to design, orchestrate, and execute end-to-end pipelines that leverage large language models to process documents, extract information, and transform data for downstream use.
Use the docetl skill to design and run a minimal pipeline that ingests a sample JSON file and prints a short summary of the results.
High-intent search queries and answers about installing and using this skill.
You can build LLM-powered data pipelines by defining YAML-based configurations for data extraction, transformation, and orchestration. This framework automates end-to-end execution for processing unstructured documents and preparing data for downstream analytics.
Extracting structured data from unstructured documents is achieved by authoring end-to-end pipelines that leverage large language models. The framework supports interactive UI playgrounds for iterative prompt engineering and schema optimization to validate extraction results.
Yes, you can define document-processing workflows using YAML-based pipeline definitions. These configurations support a modular workflow structure that includes optional scripts, references, and assets for orchestrating LLM-powered data extraction and transformation tasks.
Yes, the framework supports both interactive development and production execution. It provides an interactive UI playground for iterative prompt engineering and pipeline development, alongside a Python package for running pipelines in production analytics workflows.
You optimize prompts and schemas for LLM data extraction workflows using the interactive UI playground. This environment allows iterative prompt engineering and pipeline development to validate results and refine structured data extraction from unstructured documents.
Snowflake-LabsA concise summary of your roleplay character: name,</name> personality, and/or behavior, formatted as a single string sentence. Typically不超过50字，此处省略
starlake-aiConvert Parquet files into CSV format for batch ETL pipelines.
pradyumnasagarGuide authors through creating and validating ORS SKILL.md files.
ragnarokhaaExtract structured data from LLM responses with Pydantic validation.
