
Mcp Skill Server
Bridges the gap between sketching Python skills in your editor and running them in production. You write a script with argparse, add a SKILL.md with an entry point declaration, and the server exposes list_skills, get_skill, and run_skill tools that discover commands and parameters from your help output. The workflow is tight: edit locally in Claude Code or Cursor, use the skill immediately, then deploy the same artifact to your production MCP server without rewriting anything. Validates that your entry point is deterministic and secure before deployment. Best for teams that want skills to mature from prototype to prod without breaking the interface or execution guarantees along the way.
- GitHub stars
- 1
- tools
- 4
- transport
- STDIO
- last updated
- Feb 14, 2026
Tools
4 toolsPublic tool metadata for what this MCP can expose to an agent.
list_skillsList all available skills
List all available skills
No parameter schema in public metadata yet.
get_skillGet details about a skill (commands, parameters)
Get details about a skill (commands, parameters)
No parameter schema in public metadata yet.
run_skillExecute a skill with parameters
Execute a skill with parameters
No parameter schema in public metadata yet.
refresh_skillsReload skills after you make changes
Reload skills after you make changes
No parameter schema in public metadata yet.
Sponsors
Readme
Most coding assistants now support skills natively, so an MCP server just for skill discovery isn't necessary. Where this package adds value is making skills' execution deterministic and deployable — with a fixed entry point and controlled execution, skills developed in your editor can run in non-sandboxed production environments. It also supports incremental loading, so agents discover skills on demand instead of loading everything upfront.
MCP Skill Server
Details
- Repository
- github.com/jcc-ne/mcp-skill-server
- Package
mcp-skill-server- Transport
- STDIO
- Registry
- MCP registry, active