skills.rest
leonanpereirapinto
leonanpereirapintoCOMMUNITY
Explore all skills by this author →
𝕏@leonanpereirapGitHub

langextract

Extract structured, source-grounded data from unstructured text into JSONL outputs.

1|Updated Mar 27, 2026
Search Tags:#nlp#grounding#data-extraction#structured-data#text-processing#langextract
One-click install
npx skills add https://github.com/leonanpereirapinto/ai-utils --skill langextract-leonanpereirapinto
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: langextract
Source: https://github.com/leonanpereirapinto/ai-utils/tree/main/libs/langextract/skills/aeonbridge-langextract
Command: npx skills add https://github.com/leonanpereirapinto/ai-utils --skill langextract-leonanpereirapinto

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Extracts structured, source-grounded data from unstructured text, enabling precise traceability and easier downstream analysis across large documents.

Core Features & Use Cases

  • Long-document grounding with exact source spans for auditability.
  • Multi-model compatibility (Gemini, OpenAI, Vertex AI, and local Ollama-style setups).
  • JSONL outputs and HTML visualizations to review and validate extractions.

Quick Start

Provide a target text along with a few grounded examples and run the extraction to obtain structured results.

Frequently Asked Questions about langextract

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I extract structured data from unstructured text with source grounding?▼

To extract structured data with source grounding, you provide target text alongside grounded examples and prompt descriptions. The Skill processes the unstructured text to produce structured JSONL outputs and HTML visualizations, ensuring precise traceability for downstream analysis.

Can I use local models for long-document data extraction?▼

Yes, you can use local models for long-document data extraction. The Skill supports local Ollama-style setups alongside Gemini, OpenAI, and Vertex AI, allowing you to process extensive documents and extract source-grounded data.

What is the best way to get JSONL outputs from text processing tasks?▼

The best way to get JSONL outputs from text processing is to configure prompt descriptions and provide grounded examples before running the extraction. This generates structured JSONL files containing the extracted data mapped to exact source spans.

Does this data extraction approach work with Gemini and OpenAI?▼

Yes, this data extraction approach works with both Gemini and OpenAI. It features multi-model compatibility across Gemini, OpenAI, Vertex AI, and local models to perform source-grounded extractions from unstructured text.

How do I validate extracted structured data against original source spans?▼

You validate extracted structured data against original source spans using the generated HTML visualizations. The Skill automatically produces these visualizations alongside JSONL outputs, allowing you to review exact source-grounded traces for auditability.

Related Skills

Explore More Data & Analytics Skills→
Renzo-TognellaRenzo-Tognella

LLM Structured Outputs

Extract structured JSON from unstructured text and validate against a Pydantic schema.

Community
Advanced
RdindaRdinda
1

reversa-docs-mapper

Generate 3D and 2D architecture visualizations from cached module and dependency JSONs.

Community
Advanced
claudcherejiclaudchereji

nlp-engineer

Implement transformer-based NLP pipelines for multilingual text processing.

Community
Advanced
Fantasia1999Fantasia1999

sample-text-processor

Analyze text files with word counts, character statistics, and case conversion.

Community
Basic