Regex Entity Extractor

Overview

You give us labelled regular expressions and text; we run them on the RE2/J engine, which cannot blow up on backtracking, and hand back every match with its label, its exact value, and its character offsets, ordered by position. Up to 100 patterns per call and 1,000 matches per pattern by default. A pattern that fails to compile is reported as a warning rather than failing your whole call. Accepts file input via file_ref — no need to extract file content first.

LLMs hallucinate regex patterns and miss matches. We execute your patterns with RE2/J for linear-time, offset-annotated, complete results.

When to use it

Scope

Non-goals

• Does not perform NER or ML-based entity recognition
• Does not infer patterns — you supply every regex
• Does not validate or normalize extracted values beyond the match text

Properties

Capabilities

Cost & billing

Flat rate per invocation

File handling

File limits

Constraints this skill applies to your input and generated files. The “Limited by” column shows the static source of each cap; none of these can be raised by upgrading your plan.

How we handle your data

Input schema

Output schema

API contract

curl -sS -X POST "$BASE/v1/execute" -H "Authorization: Bearer $KEY" -H 'Content-Type: application/json' -d '{"skillKey":"text.extract_entities_regex","input":{"text":"Contact [email protected] for support.","patterns":[{"pattern":"[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}","label":"email"}]}}'

import { Squillify } from "@squillify/sdk";

const sq = new Squillify({ apiKey: process.env.SQUILLIFY_API_KEY });

const { output } = await sq.execute("text.extract_entities_regex", {"patterns":[{"pattern":"[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}","label":"email"},{"pattern":"\\+?1?[-.]?\\(?\\d{3}\\)?[-.]?\\d{3}[-.]?\\d{4}","label":"phone"}],"text":"Please contact support at [email protected] or call +1-555-867-5309 for order issues. Alternatively, reach [email protected] for invoices."});

Errors

Other limits

Limitations

• Does not perform NER or ML-based entity recognition
• Does not infer patterns — you supply every regex
• Does not validate or normalize extracted values beyond the match text
• Named entity recognition without known patterns
• Semantic entity extraction requiring context understanding
• Complex nested or recursive pattern matching beyond RE2/J limits

How to call

POST /v1/execute (requires authentication)

{
  "skillKey": "text.extract_entities_regex",
  "input": {
    "patterns": [
      {
        "pattern": "[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}",
        "label": "email"
      },
      {
        "pattern": "\\+?1?[-.]?\\(?\\d{3}\\)?[-.]?\\d{3}[-.]?\\d{4}",
        "label": "phone"
      }
    ],
    "text": "Please contact support at [email protected] or call +1-555-867-5309 for order issues. Alternatively, reach [email protected] for invoices."
  }
}

Examples

{
  "patterns": [
    {
      "pattern": "[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}",
      "label": "email"
    },
    {
      "pattern": "\\+?1?[-.]?\\(?\\d{3}\\)?[-.]?\\d{3}[-.]?\\d{4}",
      "label": "phone"
    }
  ],
  "text": "Please contact support at [email protected] or call +1-555-867-5309 for order issues. Alternatively, reach [email protected] for invoices."
}

{
  "totalMatches": 3,
  "patternCount": 2,
  "warnings": [],
  "matches": [
    {
      "label": "email",
      "value": "[email protected]",
      "start": 26,
      "end": 39
    },
    {
      "label": "phone",
      "value": "+1-555-867-5309",
      "start": 48,
      "end": 63
    },
    {
      "label": "email",
      "value": "[email protected]",
      "start": 103,
      "end": 119
    }
  ]
}

{
  "patterns": [
    {
      "pattern": "ORDER-\\d{6}",
      "label": "order_id"
    }
  ],
  "text": "2026-05-20 Processing ORDER-482917 shipped. 2026-05-21 Refund initiated for ORDER-113042."
}

{
  "totalMatches": 2,
  "patternCount": 1,
  "warnings": [],
  "matches": [
    {
      "label": "order_id",
      "value": "ORDER-482917",
      "start": 22,
      "end": 34
    },
    {
      "label": "order_id",
      "value": "ORDER-113042",
      "start": 76,
      "end": 88
    }
  ]
}

Recommended use cases

• Extract ticket numbers, order IDs, or internal codes from support text
• Pull email addresses or phone numbers from unstructured messages
• Parse log lines for timestamps, error codes, or request IDs

Avoid for

• Named entity recognition without known patterns
• Semantic entity extraction requiring context understanding
• Complex nested or recursive pattern matching beyond RE2/J limits

Related skills

Composable with

Safety & privacy

• Runs without calling any external service.
• Your payloads are never stored permanently.
• Telemetry captures metadata only — never your payload contents.
• We never train models on your payloads.
• No human reviews your payloads.

Tags