extract-version
Extract-version parses semantic version strings from various text formats and validates their compliance with semantic versioning 2.0.0 specifications. Use this skill to extract version numbers from mixed text, validate version format correctness, parse individual version components, or normalize version strings for comparison and processing.
git clone --depth 1 https://github.com/maslennikov-ig/claude-code-orchestrator-kit /tmp/extract-version && cp -r /tmp/extract-version/.claude/skills/extract-version ~/.claude/skills/extract-versionSKILL.md
# Extract Version
Parse semantic version strings and validate semver compliance.
## When to Use
- Extract version from mixed text
- Validate version format
- Parse version components
- Compare version numbers
- Normalize version strings
## Instructions
### Step 1: Receive Version Input
Accept version string in various formats.
**Expected Input**:
- `versionString`: String (e.g., "v0.7.0", "0.7.0", "version: 0.7.0", "Ver. 1.2.3-beta.1")
### Step 2: Extract Version Pattern
Use regex to extract semver pattern.
**Patterns to Match**:
- `X.Y.Z`: Basic semver
- `vX.Y.Z`: With 'v' prefix
- `X.Y.Z-prerelease`: With prerelease tag
- `X.Y.Z+build`: With build metadata
- Full semver: `X.Y.Z-prerelease+build`
**Regex**:
```
(\d+)\.(\d+)\.(\d+)(?:-([a-zA-Z0-9.-]+))?(?:\+([a-zA-Z0-9.-]+))?
```
### Step 3: Parse Components
Extract major, minor, patch, and optional components.
**Components**:
- `major`: First number
- `minor`: Second number
- `patch`: Third number
- `prerelease`: Optional prerelease tag (after -)
- `build`: Optional build metadata (after +)
### Step 4: Validate Semver
Check if version follows semantic versioning 2.0.0 spec.
**Validation Rules**:
- Major, minor, patch must be non-negative integers
- Prerelease and build must match allowed characters
- No leading zeros in numeric identifiers (except 0 itself)
### Step 5: Return Parsed Data
Return structured version object.
**Expected Output**:
```json
{
"major": 0,
"minor": 7,
"patch": 0,
"prerelease": null,
"build": null,
"original": "v0.7.0",
"normalized": "0.7.0",
"valid": true,
"semver": "0.7.0"
}
```
## Error Handling
- **No Version Found**: Return error indicating no version pattern matched
- **Invalid Format**: Return error with invalid version string
- **Invalid Component**: Describe which component is invalid
## Examples
### Example 1: Basic Version
**Input**:
```
versionString: "0.7.0"
```
**Output**:
```json
{
"major": 0,
"minor": 7,
"patch": 0,
"prerelease": null,
"build": null,
"original": "0.7.0",
"normalized": "0.7.0",
"valid": true,
"semver": "0.7.0"
}
```
### Example 2: Version with Prefix
**Input**:
```
versionString: "v1.2.3"
```
**Output**:
```json
{
"major": 1,
"minor": 2,
"patch": 3,
"prerelease": null,
"build": null,
"original": "v1.2.3",
"normalized": "1.2.3",
"valid": true,
"semver": "1.2.3"
}
```
### Example 3: Prerelease Version
**Input**:
```
versionString: "2.0.0-beta.1"
```
**Output**:
```json
{
"major": 2,
"minor": 0,
"patch": 0,
"prerelease": "beta.1",
"build": null,
"original": "2.0.0-beta.1",
"normalized": "2.0.0-beta.1",
"valid": true,
"semver": "2.0.0-beta.1"
}
```
### Example 4: Full Semver with Build
**Input**:
```
versionString: "1.0.0-alpha.1+001"
```
**Output**:
```json
{
"major": 1,
"minor": 0,
"patch": 0,
"prerelease": "alpha.1",
"build": "001",
"original": "1.0.0-alpha.1+001",
"normalized": "1.0.0-alpha.1+001",
"valid": true,
"semver": "1.0.0-alpha.1+001"
}
```
### Example 5: Extract from Text
**Input**:
```
versionString: "Version: 0.8.0 released on 2025-10-17"
```
**Output**:
```json
{
"major": 0,
"minor": 8,
"patch": 0,
"prerelease": null,
"build": null,
"original": "Version: 0.8.0 released on 2025-10-17",
"normalized": "0.8.0",
"valid": true,
"semver": "0.8.0"
}
```
### Example 6: Invalid Version
**Input**:
```
versionString: "1.2.a"
```
**Output**:
```json
{
"valid": false,
"error": "Invalid version format: patch must be numeric",
"original": "1.2.a"
}
```
## Validation
- [ ] Parses basic semver (X.Y.Z)
- [ ] Handles 'v' prefix correctly
- [ ] Extracts prerelease tags
- [ ] Extracts build metadata
- [ ] Validates semver compliance
- [ ] Extracts version from mixed text
- [ ] Returns normalized version
## Supporting Files
None required - pure parsing logic with regex.Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.
Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.
Initialize Beads issue tracking in your project with interactive configuration setup.
Calculate priority score for bugs, issues, or tasks based on severity, impact, and likelihood. Use for bug prioritization, task ordering, or risk assessment.
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
Automatically creates user-facing changelogs from git commits by analyzing commit history, categorizing changes, and transforming technical commits into clear, customer-friendly release notes. Turns hours of manual changelog writing into minutes of automated generation.
Inline orchestration workflow for dead code detection and removal with Beads integration. Provides step-by-step phases for dead-code-hunter detection, priority-based cleanup with dead-code-remover, and verification cycles.