Navi MCP Server

노출 관리 자동화를 위한 MCP 서버

문서

navi-mcp suite

An MCP server for the Tenable navi CLI (Tenable Vulnerability Management / Tenable One), plus a set of 11 companion Claude skills that document how to drive it well.

This repository is the result of a full audit and rebuild: the server's tool surface was corrected against an authoritative recursive navi --help capture, and all 11 skills were corrected to match the server + CLI and restructured for progressive disclosure.

Layout

server/      server.py — the MCP server (17 tools + resources)
skills/      the 11 corrected skills, in NAVI_SKILL_DIR layout
             (<skill>/SKILL.md, plus references/ on the denser ones)
dist/        the same 11 skills packaged as .skill files
tools/       navi_mcp_config.py — auto-detects paths, emits the install config
docs/        audit framework, gap ledger, verified findings, help-crawler
INSTALL.md   step-by-step install for Claude Desktop
README.md    this file

The 11 skills

navi (router) · navi-core · navi-mcp · navi-troubleshooting · navi-acr · navi-export · navi-scan · navi-was · navi-action · navi-explore · navi-enrich

Each SKILL.md is under 500 lines. Deep material (full schema, exhaustive selector catalog, long worked examples) lives in references/*.md and is pulled on demand.

Running the MCP server

The server shells out to the navi binary and reads the local navi.db. It does not manage API keys — set those out-of-band with navi config keys first (see skills/navi-core).

Launch it over stdio (for Claude Desktop / Claude Code) or HTTP:

python server/server.py            # stdio (default); waits for a client
python server/server.py --http     # streamable HTTP on :8000

Environment variables:

Var	Purpose	Default
`NAVI_SKILL_DIR`	Path to the `skills/` directory in this repo (so the `navi://skill/...` resources resolve)	—
`NAVI_MCP_ALLOW_WRITES`	Set to `1` to enable platform-write tools (tagging, ACR, delete, rotate, scan control, …). Off = read-only.	unset (read-only)
`NAVI_WORKDIR`	Directory holding `navi.db` and CSV exports	cwd
`NAVI_BIN`	Path to the `navi` executable	`navi`

Point NAVI_SKILL_DIR at this repo's skills/ folder (not dist/ — the server reads unpacked folders, not .skill zips).

Install in Claude Desktop

Full walkthrough in INSTALL.md. The short version: don't hand-write paths — run the helper with the Python interpreter you want Claude Desktop to use (the one that has mcp and navi), and it discovers server/server.py, your navi.db, the navi binary, and skills/, then prints (or, with --write, installs) the config:

python tools/navi_mcp_config.py            # print the mcpServers JSON
python tools/navi_mcp_config.py --write    # merge it into your Claude Desktop config (backs up first)

The launched server entry is server/server.py (use an absolute path in the config — Claude Desktop won't have your shell's PATH). After editing the config, fully quit and reopen Claude Desktop, then read navi://workdir to confirm it connected.

Tools (17) and the write-gate

Read tools (navi_explore_data, navi_explore_info, navi_explore_query SELECT, navi_export, navi_explore_api GET, scan read views, …) run freely. Platform-write tools require both NAVI_MCP_ALLOW_WRITES=1 and confirm=True and are meant to be narrated to the user before invocation: navi_enrich_tag, navi_enrich_acr, navi_enrich_add, navi_scan (create/start/stop/pause/resume), navi_was (scan/start/upload), navi_action_delete, navi_action_rotate, navi_action_cancel, navi_config(kind="url"), and navi_explore_api POST/PUT.

Resources

navi://schema/{table} — live column definitions for a navi.db table
navi://workdir — workdir, write-gate status, binary, call budget, and navi.db freshness
navi://skill/{name} — load a skill (router/core/mcp/…); lists its references
navi://skill/{name}/{ref} — load a bundled reference (e.g. navi://skill/core/schema)

Plus the navi_workflow prompt, which injects the router skill.

Long-running operations

navi exports can run for tens of minutes on large tenants — past the MCP host's ~4-minute tool-call ceiling. The server enforces a call budget (~220s) and returns a clean error naming the CLI command to run instead. Foundational syncs (navi config update full) and remote command execution (navi action push) are intentionally CLI-only. See skills/navi-core and skills/navi-troubleshooting.

Installing the skills as Claude skills

The files in dist/ are packaged for installing in Claude.ai / Claude Cowork / Claude Code as skills. (The MCP server uses the unpacked skills/ folders via NAVI_SKILL_DIR; the two are the same content in two delivery formats.)

Validation status

server.py compiles cleanly and every tool is annotated. It has not been runtime-tested against a live Tenable tenant. Before relying on it, validate the root-cause fix with one live read, e.g. navi_explore_data(subcommand="cve", cve="CVE-2021-44228"). The tool annotations require a recent mcp SDK.

See docs/verified-findings.md for the per-bug inventory and docs/gap-ledger.md for the full audit trail.