jevrecipes

Recipe catalog / text-block-role

Identify a text block’s structure

Is extracted text a heading, body paragraph, list item, code, table, caption, formula, or other block?

You have already extracted a text block but its structural markup is missing.

Explore this recipe interactively · Source and implementation guide

Use text-block-role in TypeScript

Install with npm install jev-recipes. Requires Node.js 22.9 or newer and ES modules. Set TYPESAFE_API_KEY in your server environment for live calls, which send input to TypeSafe and use API quota. See the installation guide.

import { textBlockRole } from 'jev-recipes/text-block-role';

const result = await textBlockRole({
  "text": "Installation",
  "after": "Run npm install to add the package.",
  "minConfidence": 0.8
});
console.log(result);

Input contract

FieldTypeNeeded
textstringRequired
beforestringOptional
afterstringOptional
contextstringOptional
minConfidencenumberOptional
Full input and result schemas
{
  "input": {
    "$schema": "https://json-schema.org/draft/2020-12/schema",
    "type": "object",
    "properties": {
      "text": {
        "type": "string"
      },
      "before": {
        "type": "string"
      },
      "after": {
        "type": "string"
      },
      "context": {
        "type": "string"
      },
      "minConfidence": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      }
    },
    "required": [
      "text"
    ]
  },
  "result": {
    "$schema": "https://json-schema.org/draft/2020-12/schema",
    "type": "object",
    "properties": {
      "model": {
        "type": "string"
      },
      "usage": {
        "type": "object",
        "properties": {
          "input_tokens": {
            "type": "integer",
            "minimum": 0,
            "maximum": 9007199254740991
          },
          "output_tokens": {
            "type": "integer",
            "minimum": 0,
            "maximum": 9007199254740991
          }
        },
        "required": [
          "input_tokens",
          "output_tokens"
        ],
        "additionalProperties": false
      },
      "status": {
        "type": "string",
        "enum": [
          "ready",
          "review"
        ]
      },
      "verdict": {
        "type": "string",
        "enum": [
          "heading",
          "body",
          "list_item",
          "code",
          "table",
          "caption",
          "formula",
          "other",
          "unclear"
        ]
      },
      "confidence": {
        "type": "number",
        "minimum": 0,
        "maximum": 1
      },
      "probabilities": {
        "type": "object",
        "propertyNames": {
          "type": "string",
          "enum": [
            "heading",
            "body",
            "list_item",
            "code",
            "table",
            "caption",
            "formula",
            "other",
            "unclear"
          ]
        },
        "additionalProperties": {
          "type": "number",
          "minimum": 0,
          "maximum": 1
        },
        "required": [
          "heading",
          "body",
          "list_item",
          "code",
          "table",
          "caption",
          "formula",
          "other",
          "unclear"
        ]
      }
    },
    "required": [
      "model",
      "usage",
      "status",
      "verdict",
      "confidence",
      "probabilities"
    ],
    "additionalProperties": false
  }
}

Saved example result

This hand-authored response demonstrates the contract. It is not a model accuracy measurement. Run it without an API key: npx jev-recipes demo text-block-role.

{
  "model": "demo-fixture",
  "usage": {
    "input_tokens": 0,
    "output_tokens": 0
  },
  "status": "ready",
  "verdict": "heading",
  "confidence": 0.94,
  "probabilities": {
    "heading": 0.94,
    "body": 0.04,
    "list_item": 0,
    "code": 0,
    "table": 0,
    "caption": 0,
    "formula": 0,
    "other": 0,
    "unclear": 0.02
  }
}

Evaluation evidence

Earlier-evaluator measurement

typesafe-ai/jev / 2026-09-27 / 189 held-out cases

Scoring revision 1.

82%All-case accuracy
53Sent for review
0Failed calls / cases

136 ready decisions, with 90% accuracy among those decisions.

95% case-level interval: 76% to 87%. Related synthetic cases are correlated.

Experimental: declared document acceptance policy not met

This measurement uses an earlier or unverified recipe or evaluator version. Rerun with the current recipe and evaluator before treating these numbers as current.

Full report and missesDataset and labels

Use the evaluation guide to measure this decision on your own labeled cases.

Limitations

Related recipes