[MCP Server] The Security Agent for AI assisted coding
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add vibeshift -- uv --directory path/to/cloned_repo run mcp_server.py~/.codex/config.toml
[mcp_servers.vibeshift]
command = "uv"
args = ["--directory", "path/to/cloned_repo", "run", "mcp_server.py"]
# codex mcp add vibeshift -- uv --directory path/to/cloned_repo run mcp_server.py~/.cursor/mcp.json
{
"mcpServers": {
"vibeshift": {
"command": "uv",
"args": [
"--directory",
"path/to/cloned_repo",
"run",
"mcp_server.py"
]
}
}
}.vscode/mcp.json
{
"servers": {
"vibeshift": {
"type": "stdio",
"command": "uv",
"args": [
"--directory",
"path/to/cloned_repo",
"run",
"mcp_server.py"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"vibeshift": {
"command": "uv",
"args": [
"--directory",
"path/to/cloned_repo",
"run",
"mcp_server.py"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"vibeshift": {
"command": "uv",
"args": [
"--directory",
"path/to/cloned_repo",
"run",
"mcp_server.py"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"vibeshift": {
"command": "uv",
"args": [
"--directory",
"path/to/cloned_repo",
"run",
"mcp_server.py"
]
}
}
}