Ghost Repo Context
This skill analyzes GitHub repositories for leaked credentials and sensitive data, providing valuable context to identify potential security risks.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add ghost-repo-context npx -- -y @trustedskills/ghost-repo-context
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"ghost-repo-context": {
"command": "npx",
"args": [
"-y",
"@trustedskills/ghost-repo-context"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
The ghost-repo-context skill provides AI agents with direct access to a specific Ghost CMS repository, enabling them to read, analyze, and understand the codebase without manual file uploads. It establishes a persistent context window that allows agents to navigate complex project structures and retrieve relevant files on demand.
When to use it
- Automated Code Reviews: Delegate tasks where an agent needs to scan specific files for security vulnerabilities or best practices within your Ghost theme.
- Debugging Assistance: Ask the agent to search through repository logs or configuration files to identify why a specific feature is failing.
- Documentation Generation: Instruct the agent to read existing code comments and structure files to generate up-to-date API documentation.
- Migration Planning: Have the agent analyze the current repository state to propose a strategy for migrating from one Ghost version to another.
Key capabilities
- Direct access to the contents of a designated Ghost CMS repository.
- Ability to search and retrieve specific files or directories within the codebase.
- Maintains context over multiple interactions regarding the same project structure.
Example prompts
- "Read the
config.jsfile in my Ghost repo and explain how the database connection is currently set up." - "Search the repository for any instances where custom shortcodes are defined and list them with their locations."
- "Analyze the
assets/folder to determine if there are any unused CSS files that could be removed."
Tips & gotchas
Ensure you have correctly configured the repository path in your agent settings before issuing commands, as the skill cannot access code outside of the specified scope. This feature is designed for read-only analysis; it does not execute changes or commit code back to the repository.
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.