Security scanner for third-party AI agent-skill files (SKILL.md manifests, hooks, bundled scripts). CLI, library, MCP server, and GitHub Action, all reading the same 10 rule packs, including cross-skill privilege chaining and prompt-injection detection. Zero-auth, zero-config npx or pip scan.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add skillguard -- uvx skillguard-cli~/.codex/config.toml
[mcp_servers.skillguard]
command = "uvx"
args = ["skillguard-cli"]
# codex mcp add skillguard -- uvx skillguard-cli~/.cursor/mcp.json
{
"mcpServers": {
"skillguard": {
"command": "uvx",
"args": [
"skillguard-cli"
]
}
}
}.vscode/mcp.json
{
"servers": {
"skillguard": {
"type": "stdio",
"command": "uvx",
"args": [
"skillguard-cli"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"skillguard": {
"command": "uvx",
"args": [
"skillguard-cli"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"skillguard": {
"command": "uvx",
"args": [
"skillguard-cli"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"skillguard": {
"command": "uvx",
"args": [
"skillguard-cli"
]
}
}
}