TWZRD Agent Intelligence — Claude Code MCP server: what it does
Evaluates x402 sellers and store listings before agents send payments. Works as a pre-spend inspection gate on Solana and Base.
claude mcp add --transport http twzrd-agent-intel https://intel.twzrd.xyz/mcpWhat TWZRD Agent Intelligence does
This MCP server supplies tools to inspect commerce targets before an agent signs a crypto transaction. It reviews x402 merchants against public ledger evidence to detect wash risks and assign trust scores. The advisory preflight check is free. A deeper quickCheck query costs 0.001 USDC.
Agents can also run check_listing to compare product pages and asking prices against published store evidence cards. The shopping lane is currently in pilot mode on a single item. Scored checks run on Solana and Base mainnets. Other EVM chains are recognized by the system but receive no scoring.
The service does not hold private keys and never approves payments on its own. The response parameter authorizes_spend always remains false. If running via local stdio, you must configure the TWZRD_SIGNAL_SOURCE environment variable before launch.
What it helps with
- screen x402 seller addresses before signing wallet transfers
- evaluate seller wash trading risk on Solana and Base
- compare store product URLs and prices against evidence cards
- run a zero-cost preflight check on a payment recipient
Example prompts for the agent
- Inspect the x402 seller address on Solana before approving the USDC transfer
- Run check_listing on the store URL and asking price
- Perform a free preflight risk score for the recipient on Base
How to install TWZRD Agent Intelligence
1Platform CLI: remote server
claude mcp add --transport http twzrd-agent-intel https://intel.twzrd.xyz/mcp2Platform CLI
claude mcp add --env TWZRD_SIGNAL_SOURCE=<TWZRD_SIGNAL_SOURCE> --transport stdio twzrd-agent-intelligence -- uvx twzrd-agent-intelReplace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from twzrd-agent-intel <command>
1Platform CLI: remote server
codex mcp add twzrd-agent-intelligence --url https://intel.twzrd.xyz/mcp2Platform CLI
codex mcp add twzrd-agent-intelligence --env TWZRD_SIGNAL_SOURCE=<TWZRD_SIGNAL_SOURCE> -- uvx twzrd-agent-intelReplace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from twzrd-agent-intel <command>
Config file
{
"mcpServers": {
"twzrd-agent-intelligence": {
"url": "https://intel.twzrd.xyz/mcp"
}
}
}{
"mcpServers": {
"twzrd-agent-intelligence": {
"command": "uvx",
"args": [
"twzrd-agent-intel"
],
"env": {
"TWZRD_SIGNAL_SOURCE": "<TWZRD_SIGNAL_SOURCE>"
}
}
}
}Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from twzrd-agent-intel <command>
Config file
{
"servers": {
"twzrd-agent-intelligence": {
"type": "http",
"url": "https://intel.twzrd.xyz/mcp"
}
}
}{
"servers": {
"twzrd-agent-intelligence": {
"type": "stdio",
"command": "uvx",
"args": [
"twzrd-agent-intel"
],
"env": {
"TWZRD_SIGNAL_SOURCE": "<TWZRD_SIGNAL_SOURCE>"
}
}
}
}Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from twzrd-agent-intel <command>
1Platform CLI: remote server
gemini mcp add -t http twzrd-agent-intelligence https://intel.twzrd.xyz/mcp2Platform CLI
gemini mcp add -e TWZRD_SIGNAL_SOURCE=<TWZRD_SIGNAL_SOURCE> twzrd-agent-intelligence uvx twzrd-agent-intelReplace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from twzrd-agent-intel <command>
Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more
Questions about TWZRD Agent Intelligence
How to install TWZRD Agent Intelligence in Claude Code?
Run claude mcp add --transport http twzrd-agent-intelligence https://intel.twzrd.xyz/mcp. If using stdio, run claude mcp add --env TWZRD_SIGNAL_SOURCE=<TWZRD_SIGNAL_SOURCE> --transport stdio twzrd-agent-intelligence -- uvx twzrd-agent-intel.
What does TWZRD Agent Intelligence do?
It provides inspection tools for buying agents to evaluate seller risks and catalog prices before spending. This stops automated wallets from transferring funds to unverified destinations.
Does TWZRD authorize wallet spending directly?
No, TWZRD never authorizes payments and always sets authorizes_spend to false. The agent or its local runtime retains full control over signing.
