extract_from_text

Extract structured data from unstructured text using customizable instructions and examples. Maps extracted entities to their exact source locations for precise results.

Instructions

Extract structured information from text using langextract.

Uses Large Language Models to extract structured information from unstructured text based on user-defined instructions and examples. Each extraction is mapped to its exact location in the source text for precise source grounding.

Args: text: The text to extract information from prompt_description: Clear instructions for what to extract examples: List of example extractions to guide the model config: Configuration parameters for the extraction

Returns: Dictionary containing extracted entities with source locations and metadata

Raises: ToolError: If extraction fails due to invalid parameters or API issues

Input Schema

Name	Required	Description	Default
`config`	No
`examples`	Yes
`prompt_description`	Yes
`text`	Yes

Input Schema (JSON Schema)

{
  "$defs": {
    "ExtractionConfig": {
      "description": "Configuration for extraction parameters.",
      "properties": {
        "extraction_passes": {
          "default": 1,
          "description": "Number of extraction passes for better recall",
          "title": "Extraction Passes",
          "type": "integer"
        },
        "max_char_buffer": {
          "default": 1000,
          "description": "Max characters per chunk",
          "title": "Max Char Buffer",
          "type": "integer"
        },
        "max_workers": {
          "default": 10,
          "description": "Max parallel workers",
          "title": "Max Workers",
          "type": "integer"
        },
        "model_id": {
          "default": "gemini-2.5-flash",
          "description": "LLM model to use",
          "title": "Model Id",
          "type": "string"
        },
        "temperature": {
          "default": 0.5,
          "description": "Sampling temperature (0.0-1.0)",
          "title": "Temperature",
          "type": "number"
        }
      },
      "title": "ExtractionConfig",
      "type": "object"
    },
    "ExtractionExample": {
      "description": "Model for extraction examples.",
      "properties": {
        "extractions": {
          "description": "Expected extractions",
          "items": {
            "additionalProperties": true,
            "type": "object"
          },
          "title": "Extractions",
          "type": "array"
        },
        "text": {
          "description": "Example text",
          "title": "Text",
          "type": "string"
        }
      },
      "required": [
        "text",
        "extractions"
      ],
      "title": "ExtractionExample",
      "type": "object"
    }
  },
  "properties": {
    "config": {
      "$ref": "#/$defs/ExtractionConfig",
      "default": {
        "extraction_passes": 1,
        "max_char_buffer": 1000,
        "max_workers": 10,
        "model_id": "gemini-2.5-flash",
        "temperature": 0.5
      },
      "title": "Config"
    },
    "examples": {
      "items": {
        "$ref": "#/$defs/ExtractionExample"
      },
      "title": "Examples",
      "type": "array"
    },
    "prompt_description": {
      "title": "Prompt Description",
      "type": "string"
    },
    "text": {
      "title": "Text",
      "type": "string"
    }
  },
  "required": [
    "text",
    "prompt_description",
    "examples"
  ],
  "type": "object"
}

LangExtract MCP Server