Explainable ALLOW/WARN/BLOCK risk decisions for AI agent actions before execution, signed as OAA.
Setup
Choose your client, copy the config, then fill in any required environment variables.
claude mcp add agentic-wallet-guardian-v3 -- uvx agentic-wallet-guardian-mcp~/.codex/config.toml
[mcp_servers.agentic-wallet-guardian-v3]
command = "uvx"
args = ["agentic-wallet-guardian-mcp"]
# codex mcp add agentic-wallet-guardian-v3 -- uvx agentic-wallet-guardian-mcp~/.cursor/mcp.json
{
"mcpServers": {
"agentic-wallet-guardian-v3": {
"command": "uvx",
"args": [
"agentic-wallet-guardian-mcp"
]
}
}
}.vscode/mcp.json
{
"servers": {
"agentic-wallet-guardian-v3": {
"type": "stdio",
"command": "uvx",
"args": [
"agentic-wallet-guardian-mcp"
]
}
}
}~/.gemini/settings.json
{
"mcpServers": {
"agentic-wallet-guardian-v3": {
"command": "uvx",
"args": [
"agentic-wallet-guardian-mcp"
]
}
}
}claude_desktop_config.json
{
"mcpServers": {
"agentic-wallet-guardian-v3": {
"command": "uvx",
"args": [
"agentic-wallet-guardian-mcp"
]
}
}
}.trae/mcp.json
{
"mcpServers": {
"agentic-wallet-guardian-v3": {
"command": "uvx",
"args": [
"agentic-wallet-guardian-mcp"
]
}
}
}