Adversarial multi-model reasoning verification for AI agents before trades execute. Claude, Grok, and DeepSeek challenge each decision — returns ALLOW or HOLD with JWKS-signed attestation. x402-gated on Base (USDC). Part of the 4-issuer Combined Attestation Standard with InsumerAPI, RNWY, and Maiat.
Run the underlying MCP server directly — without Agent-CoreX routing.
npx -y thoughtproof-mcpAgent-CoreX intelligently routes which tools from thoughtproof/thoughtproof-mcp your AI agent actually needs per request — reducing token usage and cost.
agent-corex.json
{
"mcpServers": {
"thoughtproof-mcp": {
"command": "npx",
"args": [
"-y",
"thoughtproof-mcp"
]
}
}
}Connect in 2 minutes. Only pay for the tools your agent actually uses.
Recommended: intelligent tool routing, lower costs
uvx agent-corex mcp add thoughtproof-mcpAgent-CoreX selects only the tools your agent needs, cutting token usage by up to 60%.
Use thoughtproof/thoughtproof-mcp directly without routing
npx -y thoughtproof-mcpNew to Agent-CoreX? View setup guide →