The Engine

A Deterministic Agentic Engine and CLI for Exposure Management.

Most AI agents guess. Navi doesn't — every answer and action is a real, reproducible tool call against your data. Ask the same question twice, get the same answer. No hallucinated finding, no fabricated asset. Built on Tenable. Grounded in your data. Reproducible by design.

The Four Pillars

01

THE ENGINE

The navi CLI — open-source Python that syncs Tenable VM, Tenable One, and Tenable WAS into a local navi.db you query in milliseconds. Every hound hunts on data navi fetched, normalized, and indexed.

02

THE MCP SERVER

navi-mcp exposes 20 typed tools and live schema resources to any MCP-capable AI. Reads run free; writes are double-gated — server flag plus per-call confirm. Email and remote-exec each need their own extra flag on top.

03

THE SKILLS

17 Claude skills act as guardrails — a router plus 15 domain skills covering every navi domain, plus the Hounds playbook skill that grounds each hound's detection logic, tag scheme, and selectors. Written once, followed every time.

04

THE HOUNDS

41 specialised agents — each hunting one exposure class. Together they cover the full attack surface: KEV, attack paths, identity, certificates, post-quantum crypto, AI, IoT, software, compliance, web apps, ownership, remediation, and more.

By the Numbers

353K+Total downloads on PyPI
98★ / 34GitHub stars / forks
2M+Assets in the largest tested accounts
1TB → DockerScales from terabyte databases down to nimble containers

Download stats from pepy.tech · pip install navi-pro · GPLv3 licensed

Open Source

Free. Open. Battle-tested since 2019.

■ navi-mcp and 17 Claude skills extend navi into the agentic layer the Hounds run on. navi is GPLv3; navi-mcp is MIT. pip install navi-pro → navi config keys → navi config update full.