Skill Vision Control (SVC)
A version management server for MCP skills that enables automated update detection, parallel version testing, and smart merging of custom modifications with official updates. It allows users to safely switch, rollback, and manage local skill versions through a standardized suite of MCP tools.
README
Skill Vision Control (SVC)
Safe MCP Skill Version Manager - Detect updates, parallel testing, smart merge, confirm before replace
Features
- ๐ Update Detection - Automatically detect new versions from GitHub/npm
- ๐ฆ Version Management - Keep multiple versions, switch anytime
- ๐ Smart Merge - Merge official updates with your custom changes
- ๐งช A/B Testing - Test new versions before switching
- โฐ Scheduled Checks - Automatic weekly/monthly update checks
- ๐ Notifications - Desktop notifications for updates
- ๐ค MCP Server - Let AI manage your skills
Installation
npm install -g skill-vision-control
Or with yarn:
yarn global add skill-vision-control
Quick Start
# Add a skill to manage
svc add weather --source github:username/weather-mcp
# Check for updates
svc check
# Download new version (keeps old version)
svc download weather
# Test and switch
svc switch weather --version v1.1.0
# Or if you have custom changes, merge them
svc merge weather
Commands
Skill Management
| Command | Description |
|---|---|
svc add <name> --source <url> |
Register a skill (github:user/repo or npm:package) |
svc list |
List all managed skills |
svc info <name> |
Show detailed information |
svc remove <name> |
Remove a skill |
Version Control
| Command | Description |
|---|---|
svc check [name] |
Check for updates |
svc download <name> |
Download new version (keep old) |
svc versions <name> |
List all local versions |
svc switch <name> -v <version> |
Switch to specific version |
svc rollback <name> |
Rollback to previous version |
svc confirm <name> |
Confirm current version |
svc cleanup <name> --keep <n> |
Clean old versions |
Custom Modifications
| Command | Description |
|---|---|
svc fork <name> |
Create custom branch for modifications |
svc save <name> -c "comment" |
Save your modifications |
svc diff <name> |
View differences from official |
svc merge <name> |
Merge official update with your changes |
svc conflicts <name> |
View merge conflicts |
svc resolve <name> -f <file> -u <choice> |
Resolve conflicts |
Schedule
| Command | Description |
|---|---|
svc schedule set -i <days> |
Set check interval (1/7/14/30 days) |
svc schedule show |
Show current schedule |
svc schedule enable |
Enable scheduled checks |
svc schedule disable |
Disable scheduled checks |
svc schedule run |
Manually trigger check |
Workflow Examples
Basic Update Flow
# 1. Check for updates
svc check
# Output: weather: v1.0.0 โ v1.1.0 available
# 2. Download (old version preserved)
svc download weather
# 3. Test new version
svc switch weather -v v1.1.0 -t official
# 4. If good, confirm; if not, rollback
svc confirm weather
# or
svc rollback weather
Custom Changes + Update
# 1. Create custom branch
svc fork weather
# 2. Make your modifications...
# 3. Save changes
svc save weather -c "Added Chinese language support"
# 4. Later, when update available
svc check
# Output: โ ๏ธ You have custom changes. Use "svc merge"
# 5. Download and merge
svc download weather
svc merge weather
# 6. If conflicts exist
svc conflicts weather
svc resolve weather -f src/config.ts -u custom
# 7. Test merged version
svc switch weather -v v1.1.0-merged -t merged
# 8. Confirm
svc confirm weather
Using as MCP Server
Add to your MCP configuration:
{
"mcpServers": {
"skill-vision-control": {
"command": "svc",
"args": ["serve"]
}
}
}
Available MCP tools:
svc_list_skills- List all managed skillssvc_get_skill_info- Get skill detailssvc_check_updates- Check for updatessvc_get_versions- Get local versionssvc_switch_version- Switch versionsvc_rollback- Rollback to previoussvc_download_update- Download new versionsvc_merge- Merge with custom changessvc_get_conflicts- View merge conflicts
Data Storage
All data is stored in ~/.svc/:
~/.svc/
โโโ skills.json # Skill registry
โโโ schedule.json # Schedule settings
โโโ config.json # Global config
โโโ versions/ # Version storage
โโโ <skill>/
โโโ official/
โโโ custom/
โโโ merged/
โโโ active -> ...
Configuration
Supported Sources
- GitHub:
github:username/repoorusername/repo - npm:
npm:package-name
Schedule Options
1d- Daily checks7d- Weekly checks (default)14d- Bi-weekly checks30d- Monthly checks
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
MIT License - see LICENSE for details.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
E2B
Using MCP to run code via e2b.