regex-tester-mcp

Regex Tester MCP server. Test, explain, and template regular expressions.

No shell. No external tools. Pure native RegExp.

Install

npm install -g regex-tester-mcp

Usage

Claude Desktop / Cursor / OpenCode

Add to your MCP config:

{
  "mcpServers": {
    "regex-tester": {
      "command": "npx",
      "args": ["-y", "regex-tester-mcp"]
    }
  }
}

Tools

test_regex — Test Regex

Test a regular expression against a string. Returns all matches with capture groups.

Input:

{
  "pattern": "(\\d+)",
  "test_string": "abc123def456",
  "flags": "g"
}

Output:

{
  "pattern": "(\\d+)",
  "flags": "g",
  "test_string": "abc123def456",
  "match_count": 2,
  "matches": [
    { "match": "123", "groups": { "1": "123" }, "index": 3 },
    { "match": "456", "groups": { "1": "456" }, "index": 9 }
  ]
}

explain_regex — Explain Regex

Explain a regular expression in human-readable form.

Input:

{
  "pattern": "^\\d{3}-\\d{2}-\\d{4}$"
}

Output:

Regular expression: /^\d{3}-\d{2}-\d{4}$/

Breakdown:
  1. '^' — Matches the start of the string
  2. '\d' — Matches any digit (equivalent to [0-9])
  3. '{3}' — Quantifier: Matches exactly 3 times
  4. '-' — Literal character: '-'
  5. '\d' — Matches any digit (equivalent to [0-9])
  6. '{2}' — Quantifier: Matches exactly 2 times
  7. '-' — Literal character: '-'
  8. '\d' — Matches any digit (equivalent to [0-9])
  9. '{4}' — Quantifier: Matches exactly 4 times
  10. '$' — Matches the end of the string

regex_templates — Regex Templates

Get common regex templates. Optionally filter by category.

Input:

{
  "category": "validation"
}

Output:

{
  "count": 10,
  "category": "validation",
  "templates": [
    {
      "name": "email",
      "category": "validation",
      "pattern": "^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}$",
      "description": "Email address",
      "example": "[email protected]"
    }
  ]
}

Built-in Templates

| Name | Category | Description | |------|----------|-------------| | email | validation | Email address | | url | validation | HTTP/HTTPS URL | | phone_tw | validation | Taiwan phone number | | phone_us | validation | US phone number | | ipv4 | validation | IPv4 address | | date_iso | validation | ISO date (YYYY-MM-DD) | | time_24h | validation | 24-hour time (HH:MM) | | hex_color | validation | Hex color code | | credit_card | validation | Credit card (Visa/MC/Amex) | | uuid | validation | UUID v4 | | html_tag | extraction | HTML tag extraction | | digits | extraction | Extract all digits | | words | extraction | Extract all words | | hashtags | extraction | Extract hashtags |

Design

| Feature | Why | |---------|-----| | Zero runtime deps | Only @modelcontextprotocol/sdk and zod | | Native RegExp | No external regex libraries needed | | Structured output | AI reads matches/groups precisely | | Human-readable explanations | Token-by-token breakdown | | 14 built-in templates | Common patterns ready to use |

License

MIT