# LangExtract Web

> Use this tool when you need to extract structured information from text using large language models, solving problems such as data mining, text analysis, and information retrieval. It takes in unstructured text as input and outputs extracted data with precise source grounding, supporting few-shot learning and long document optimization. Ideal for use cases requiring advanced natural language processing, LangExtract Web provides a flexible interface through its web UI, REST API, or MCP server.

Canonical page: https://skillsregistry.net/skills/neosun100-langextract-web  
JSON: https://api.skillsregistry.net/v1/skills/neosun100-langextract-web

## Description

LangExtract Web provides a containerized interface to Google's LangExtract library for extracting structured information from text using large language models. It offers three access methods: a modern web UI with dark mode support, a REST API with Swagger documentation, and an MCP server for AI assistant integration. The tool supports precise source grounding where extractions map back to exact text positions, few-shot learning, long document optimization through chunking, and multiple LLM providers including Gemini, OpenAI, Claude, and Ollama via LiteLLM.

## Trust

- **Trust score (0–1):** 0.33
- **Verification tier:** scanned
- **Last scanned:** 2026-09-28

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** ai-ml
- **Updated:** 2026-09-28

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/neosun100-langextract-web)
- **Repository:** <https://github.com/neosun100/langextract-web>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "neosun100-langextract-web"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/neosun100-langextract-web` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/neosun100-langextract-web/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt