Jeevesus
Natural-language threat-intel search. Type a question in English — get IOCs, adversaries, blog posts, Epstein-files matches, STIX bundles. 17.9M documents. Free tier, 25 queries/day, no auth gate.
Two read-only MCP servers on the official registry, backed by the same brain. Jeevesus answers questions about threats. Dredd refuses to invoke anything compromised.
Natural-language threat-intel search. Type a question in English — get IOCs, adversaries, blog posts, Epstein-files matches, STIX bundles. 17.9M documents. Free tier, 25 queries/day, no auth gate.
Pre-flight security check for the MCP ecosystem. Blocks calls against known-compromised dependencies, tool-surface drift, remote-URL hijacks, and permission escalation. HMAC-signed verdicts in <200ms.
Same threat corpus, surfaced wherever you actually live — your editor, your SIEM, your terminal, your shell, your browser, your CI/CD pipeline. Open-source. MIT-licensed. Receipts on each repo.
We measure how AI models see brands — and we use AI models to measure ourselves. Customer enrichment, trust signals, AIPM scoring all run through a four-model quorum so no single model's bias becomes truth.
Drop one of these into your MCP client config. Restart. You'll see Dredd available with one tool: check_mcp_server.
{
"mcpServers": {
"dredd": {
"url": "https://analytics.dugganusa.com/api/v1/dredd/mcp"
}
}
}
{
"mcpServers": {
"dredd": {
"url": "https://analytics.dugganusa.com/api/v1/dredd/mcp"
}
}
}
curl -X POST \
https://analytics.dugganusa.com/api/v1/dredd/mcp \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}'