The trust score for AI-generated code. One command verifies what your coding agent shipped: security findings with file:line evidence, Docker-sandboxed build/boot checks, and claim receipts that catch agents lying about tests.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add umbra -- npx -y @elberacasa/umbra~/.codex/config.toml
[mcp_servers.umbra]
command = "npx"
args = ["-y", "@elberacasa/umbra"]
# codex mcp add umbra -- npx -y @elberacasa/umbra~/.cursor/mcp.json
{
"mcpServers": {
"umbra": {
"command": "npx",
"args": [
"-y",
"@elberacasa/umbra"
]
}
}
}.vscode/mcp.json
{
"servers": {
"umbra": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@elberacasa/umbra"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"umbra": {
"command": "npx",
"args": [
"-y",
"@elberacasa/umbra"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"umbra": {
"command": "npx",
"args": [
"-y",
"@elberacasa/umbra"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"umbra": {
"command": "npx",
"args": [
"-y",
"@elberacasa/umbra"
]
}
}
}