Skill Jack MCP
An MCP server that jacks Agent Skills directly into your LLM's brain.
Features
Skill Discovery - Automatically finds skills in a configured directory
Server Instructions - Injects skill metadata into the client's system prompt
Skill Tool - Load full skill content on demand (progressive disclosure)
Installation
Usage
Windows note: Use forward slashes in paths when using with MCP Inspector:
How It Works
The server implements the Agent Skills progressive disclosure pattern:
At startup: Discovers all skills and generates
<available_skills>XML for server instructionsOn connection: Client receives skill metadata (name, description, location) in system prompt
On tool call: Agent calls
skilltool to load full SKILL.md content
Tools
skill
Load and activate an Agent Skill by name. Returns the full SKILL.md content.
Input:
Output: Full SKILL.md content including frontmatter and instructions.
skill-resource
Read files within a skill's directory (scripts/, references/, assets/, snippets/, etc.).
This follows the Agent Skills spec's progressive disclosure pattern - resources are loaded only when needed.
Input:
Output: File content.
List available files (pass empty path):
Security: Path traversal is prevented - only files within the skill directory can be accessed.
Security
Skills are treated as trusted content. This server reads and serves skill files directly to clients without sanitization. Only configure skills directories containing content you trust.
Protections in place:
Path traversal prevention (symlink-aware)
File size limits (10MB max)
Directory depth limits
Skill content is confined to configured directories
Not protected against:
Malicious content within trusted skill directories
Prompt injection via skill instructions (skills can influence LLM behavior by design)
Server Instructions Format
The server generates instructions that include a usage preamble and skill metadata:
This follows MCP server instructions best practices:
Concise, actionable guidance
No redundancy with tool descriptions
Focused on workflow relationships
Skill Discovery
The server scans the provided directory for subdirectories containing a SKILL.md file:
Each SKILL.md must have valid YAML frontmatter with name and description fields.
Testing
Configuration for Claude Desktop
Add to your claude_desktop_config.json: