Claude Regex Pattern Builder & Explainer Prompt
You are a regex expert who can read and write complex patterns and explain them clearly to any developer.
Category
💻 Coding
Difficulty
Intermediate
Models
3
Last Updated
2026-06-28
Works with
📄 Example output
⚠️ Common Mistakes
❓ FAQ
⚙️ Fill in your variables
📋 Prompt
You are a regex expert who can read and write complex patterns and explain them clearly to any developer.
What to match: [what to match]
Should match: [examples that should match — at least 3]
Should NOT match: [examples that should NOT match — at least 3]
Language: [programming language — Python/JavaScript/Go/PHP/Java]
Task:
1. THE PATTERN: Complete regex with flags, ready to use
2. VISUAL BREAKDOWN: Each part of the regex explained with ←── labels
3. CODE SNIPPET: Working code in [language] testing all your examples
4. EDGE CASES: What the pattern will and won't handle
5. ALTERNATIVE PATTERN: A simpler version if the full pattern is overkill
6. COMMON MISTAKES: Pitfalls for this type of pattern
What to match: [what to match]
Should match: [examples that should match — at least 3]
Should NOT match: [examples that should NOT match — at least 3]
Language: [programming language — Python/JavaScript/Go/PHP/Java]
Task:
1. THE PATTERN: Complete regex with flags, ready to use
2. VISUAL BREAKDOWN: Each part of the regex explained with ←── labels
3. CODE SNIPPET: Working code in [language] testing all your examples
4. EDGE CASES: What the pattern will and won't handle
5. ALTERNATIVE PATTERN: A simpler version if the full pattern is overkill
6. COMMON MISTAKES: Pitfalls for this type of pattern
GOAL: Match valid email addresses
Should match: [email protected], [email protected], [email protected]
Should NOT match: user@, @domain.com, user @domain.com, [email protected]
PATTERN (Python):
```python
import re
EMAIL_PATTERN = re.compile(
r'^[\w.+\-]+@[\w\-]+(\.[\w\-]+)*\.[a-zA-Z]{2,}$',
re.IGNORECASE
)
```
VISUAL BREAKDOWN:
```
^ ← start of string
[\w.+\-]+ ← local part: word chars, dots, plus, hyphens (1+ chars)
@ ← literal @ symbol
[\w\-]+ ← domain name (word chars and hyphens)
(\.[\w\-]+)* ← optional subdomains (.subdomain repeated 0+ times)
\. ← dot before TLD (escaped)
[a-zA-Z]{2,} ← TLD: 2+ letters (com, co, io, museum...)
$ ← end of string
```
CODE TEST:
```python
test_cases = [
('[email protected]', True),
('[email protected]', True),
('user@', False),
('@domain.com', False),
]
for email, expected in test_cases:
result = bool(EMAIL_PATTERN.match(email))
status = '✅' if result == expected else '❌'
print(f'{status} {email}: {result}')
```
EDGE CASES:
✅ Handles: subdomains, plus addressing, international TLDs
❌ Does NOT handle: IP address email hosts (user@[192.168.1.1]), Unicode domains
Note: RFC 5321-compliant email validation requires a parser, not just regex
Should match: [email protected], [email protected], [email protected]
Should NOT match: user@, @domain.com, user @domain.com, [email protected]
PATTERN (Python):
```python
import re
EMAIL_PATTERN = re.compile(
r'^[\w.+\-]+@[\w\-]+(\.[\w\-]+)*\.[a-zA-Z]{2,}$',
re.IGNORECASE
)
```
VISUAL BREAKDOWN:
```
^ ← start of string
[\w.+\-]+ ← local part: word chars, dots, plus, hyphens (1+ chars)
@ ← literal @ symbol
[\w\-]+ ← domain name (word chars and hyphens)
(\.[\w\-]+)* ← optional subdomains (.subdomain repeated 0+ times)
\. ← dot before TLD (escaped)
[a-zA-Z]{2,} ← TLD: 2+ letters (com, co, io, museum...)
$ ← end of string
```
CODE TEST:
```python
test_cases = [
('[email protected]', True),
('[email protected]', True),
('user@', False),
('@domain.com', False),
]
for email, expected in test_cases:
result = bool(EMAIL_PATTERN.match(email))
status = '✅' if result == expected else '❌'
print(f'{status} {email}: {result}')
```
EDGE CASES:
✅ Handles: subdomains, plus addressing, international TLDs
❌ Does NOT handle: IP address email hosts (user@[192.168.1.1]), Unicode domains
Note: RFC 5321-compliant email validation requires a parser, not just regex
🏆
💡 Pro Tips
Best model for this prompt
DeepSeek
DeepSeek V3 / R1
Test your regex against both positive AND negative examples before shipping
Use named groups (?P<name>...) in Python for readability in complex patterns
Regex is greedy by default — add ? after quantifiers (.*? ) for non-greedy matching when needed
For production email validation, use a proper validation library — regex only gets you 95% correct
Using .* when .+? or a specific character class would be safer and more precise
Forgetting to escape special characters — . * + ? [ ] { } ( ) ^ $ | \ all have special meaning
Not anchoring with ^ and $ — without anchors, partial matches succeed when full-string matching is needed
Catastrophic backtracking — nested quantifiers like (a+)+ on mismatched input can freeze your application
- When should I NOT use regex?For parsing HTML/XML (use a proper parser), complex natural language, full email validation per RFC, or URL parsing. Regex is powerful but has fundamental limitations on context-free and context-sensitive grammars.
- How do I test regex quickly?regex101.com is the best free tool — paste your pattern, test string, and get visual highlighting plus explanation of each match group. Language-specific flavour matters (Python re vs JavaScript RegExp differ slightly).
- Is DeepSeek good at regex?Yes — DeepSeek V3 and R1 are strong at regex. R1's reasoning mode is particularly useful for complex patterns because it thinks through the edge cases before writing the pattern.
- What's the difference between regex flavours?Python (re module): PCRE-like. JavaScript: no lookbehind in older engines. Go: RE2 (no backreferences). PHP: PCRE. Java: java.util.regex. The core syntax is similar but advanced features (lookbehind, named groups, possessive quantifiers) vary.