Skill Create
Sourman's skill-create generates custom AI agent skills based on user prompts, defining purpose, inputs, and outputs.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add skill-create npx -- -y @trustedskills/skill-create
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"skill-create": {
"command": "npx",
"args": [
"-y",
"@trustedskills/skill-create"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
skill-create
What it does
This foundational agent capability enables the creation of new skills, allowing users to define custom functionalities and extend an AI agent's operational scope. It serves as the primary mechanism for building reusable logic blocks that agents can execute to perform specific tasks.
When to use it
- You need to add a specialized function to your agent that isn't covered by default capabilities.
- You want to standardize a recurring task across multiple different AI workflows.
- You are developing a custom toolset for an enterprise environment with unique compliance or data handling requirements.
- You wish to modularize complex logic into smaller, testable units for better maintainability.
Key capabilities
- Defining new functional blocks for AI agents
- Extending standard agent behavior with custom logic
- Creating reusable components for workflow automation
- Establishing the foundation for specialized task execution
Example prompts
- "Create a skill that summarizes long documents into bullet points while preserving key dates and names."
- "Define a new capability to automatically format raw JSON data into readable Markdown tables."
- "Build a custom tool that cross-references user input against a specific internal knowledge base and returns relevant citations."
Tips & gotchas
Ensure you clearly define the input parameters and expected output format for any skill you create to prevent ambiguity during execution. Test new skills in a sandbox environment first to verify they integrate correctly with your agent's existing workflow before deploying them broadly.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.