MCP Observatory
5m ago Tracked 26,551 Last 5m ago 24h 1,897
github Python analyzed 1786ad9

homeassistant-ai/ha-mcp

github

The Unofficial and Awesome Home Assistant MCP Server

maintainer
homeassistant-ai
license
MIT
first seen
2026-06-22
last seen
2026-08-01
releases · 30d
192
short id
risk 87/100 · heuristic grade
E critical
  • vulnerabilities attested + 35
  • capability exposure inferred + 35
  • recent drift inferred + 20
  • tool safety inferred + 5
  • trust mitigators mixed − 8

attested inferred mixed

The A–E grade is our heuristic synthesis — a "review this" prompt, not a verdict. Each factor is tagged by what backs it: attested (a verifiable record), reported (a third party's claim), or inferred (our own heuristic, e.g. permissions). See methodology.

graded 6m ago · see ecosystem CVEs →

risk trajectory 4 movements
  • E · 80 → E · 87
  • D · 72 → E · 80
  • B · 20 → D · 72
  • A · 0 → B · 20
capability exposure grade factor +35
Inferred surface — each links to servers holding it:
vulnerabilities 5 CVEs · grade factor +35
MEDIUM
OAuth-mode Settings Routes Missing Authentication affects ["<= 7.13"]
MEDIUM
Add-on settings and policy routes are reachable without authentication at the bare root path affects ["<= 7.6.0"]
MEDIUM
YAML config backups written under www/ are served unauthenticated at /local/ affects ["< 7.4.1.dev456"]
MEDIUM
XSS via Unescaped HTML in OAuth Consent Form EPSS 0% CVE-2026-32112 affects ["<=6.7.2"]
MEDIUM
OAuth 2.1 DCR mode enables network reconnaissance via an error oracle EPSS 0% CVE-2026-32111 affects ["<=6.7.2"]
tool safety 1 findings · grade factor +5
  1. medium dangerous code

    dynamic exec: __import__(), eval()/exec()

skills & danger signals github-tarball
prompt-surface shipped agent-instruction files + hidden-content / dangerous-code findings — quoted from the analyzed source

analyzed commit 1786ad9 · analyzer v28 · 2d ago

skills & prompt files 16

danger signals3

other grade factors evidence elsewhere
embed badge readme-ready
live risk-grade badge preview [![MCP Observatory risk grade](https://mcpobservatory.com/servers/github:homeassistant-ai/ha-mcp/badge.svg)](https://mcpobservatory.com/servers/github:homeassistant-ai/ha-mcp/security)

Heuristic, inferred signals — false positives (legitimately powerful tools, forks, language ports) are expected. Treat each as "review this", not a verdict. See the ecosystem-wide picture on the security hub, or the fleet security of homeassistant-ai.