agentshield mcp for Grok
io.github.dl-eigenart/agentshield-mcp
Detect prompt injection, jailbreak, and social-engineering attacks in LLM agents.
client:Grok
transport:stdio
runtime:npm
Install agentshield mcp in Grok
.mcp.json (in your project root)
{
"mcpServers": {
"agentshield-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@eigenart/agentshield-mcp"
],
"env": {
"AGENTSHIELD_API_KEY": "<AGENTSHIELD_API_KEY>"
}
}
}
}
Grok Build reads Claude Code's .mcp.json and Claude Desktop's config unchanged, so a server you have already set up there needs nothing new. The Grok web app and X integration do not take custom servers. Grok docs