regex-explainer
Explains regular expressions in plain language, builds new patterns from examples and tests them against matching and non-matching samples. Use when the user pastes a regex they do not understand, needs a pattern for a text format, or wants to find out why a pattern does not matc
Install
npx skills add https://github.com/Lethe044/agentskill-kit/tree/main/skills/regex-explainer
claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install lethe044-agentskill-kit@llmmart
git clone https://github.com/Lethe044/agentskill-kit.git
The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole lethe044/agentskill-kit collection as a plugin from our marketplace. Git is the plain clone.
Skill manifest
Regex explainer
Regular expressions are dense. Read them in pieces and prove the result with examples.
When to use
- The user pastes a pattern and asks what it does.
- The user describes a format (dates, ids, log lines) and needs a pattern.
- A pattern matches too much, too little, or is very slow.
Explaining a pattern
- Name the flavor (Python
re, JavaScript, PCRE, POSIX). Behavior differs for lookbehind, named groups and flags. - Split the pattern into tokens: anchors, groups, character classes, quantifiers, alternations.
- Explain each token in order, in a short list, then give the whole meaning in one sentence.
- Show two strings that match and two that do not, and say why.
Building a pattern
- Ask for at least three strings that must match and three that must not, unless they are given.
- Start from the simplest pattern that matches the positives, then tighten it with anchors (
^,$,\b) until the negatives fail. - Prefer explicit character classes (
[0-9],[A-Za-z]) over.and avoid.*in the middle of a pattern. - Use non-capturing groups
(?:...)when you do not need the text, and named groups when you do. - Test the pattern against every example and show a table of results. If a code runner is available, run it.
Common problems
- Greedy quantifiers swallow too much. Use a lazy quantifier (
.*?) or a negated class ([^"]*). - Missing anchors make a pattern match inside longer text. Add
^...$or\b. - Unescaped special characters (
.,?,+,() inside literal text. - Nested quantifiers such as
(a+)+can take exponential time on long input. Rewrite them. - Do not use a regex to parse nested or recursive formats (HTML, JSON, source code). Use a parser.
Example
Pattern: ^(\d{4})-(\d{2})-(\d{2})$
Matches a date like 2024-05-17 and captures year, month and day. It rejects 2024-5-17 (month has one digit)
and x2024-05-17 (anchored). It does not check that the month is 1 to 12.
Errors
If the flavor is unknown and the pattern uses flavor specific syntax, say which flavors support it.
Files (agentskill-kit)
-
evals
-
evals.json 399 B
Loading…
-
-
SKILL.md 2.5 KB
--- name: regex-explainer description: Explains regular expressions in plain language, builds new patterns from examples and tests them against matching and non-matching samples. Use when the user pastes a regex they do not understand, needs a pattern for a text format, or wants to find out why a pattern does not match. license: MIT metadata: version: 1.0.0 author: Lethe044 tags: [regex, text, parsing, explain] --- # Regex explainer Regular expressions are dense. Read them in pieces and prove the result with examples. ## When to use - The user pastes a pattern and asks what it does. - The user describes a format (dates, ids, log lines) and needs a pattern. - A pattern matches too much, too little, or is very slow. ## Explaining a pattern 1. Name the flavor (Python `re`, JavaScript, PCRE, POSIX). Behavior differs for lookbehind, named groups and flags. 2. Split the pattern into tokens: anchors, groups, character classes, quantifiers, alternations. 3. Explain each token in order, in a short list, then give the whole meaning in one sentence. 4. Show two strings that match and two that do not, and say why. ## Building a pattern 1. Ask for at least three strings that must match and three that must not, unless they are given. 2. Start from the simplest pattern that matches the positives, then tighten it with anchors (`^`, `$`, `\b`) until the negatives fail. 3. Prefer explicit character classes (`[0-9]`, `[A-Za-z]`) over `.` and avoid `.*` in the middle of a pattern. 4. Use non-capturing groups `(?:...)` when you do not need the text, and named groups when you do. 5. Test the pattern against every example and show a table of results. If a code runner is available, run it. ## Common problems - Greedy quantifiers swallow too much. Use a lazy quantifier (`.*?`) or a negated class (`[^"]*`). - Missing anchors make a pattern match inside longer text. Add `^...$` or `\b`. - Unescaped special characters (`.`, `?`, `+`, `(`) inside literal text. - Nested quantifiers such as `(a+)+` can take exponential time on long input. Rewrite them. - Do not use a regex to parse nested or recursive formats (HTML, JSON, source code). Use a parser. ## Example Pattern: `^(\d{4})-(\d{2})-(\d{2})$` Matches a date like `2024-05-17` and captures year, month and day. It rejects `2024-5-17` (month has one digit) and `x2024-05-17` (anchored). It does not check that the month is 1 to 12. ## Errors If the flavor is unknown and the pattern uses flavor specific syntax, say which flavors support it.
Comments (0)
Sign in to join the conversation.
Reviews (0)
No reviews yet.
No comments yet.