GHSA-q855-8rh5-jfgq

Suggest an improvement
Source
https://github.com/advisories/GHSA-q855-8rh5-jfgq
Import Source
https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/07/GHSA-q855-8rh5-jfgq/GHSA-q855-8rh5-jfgq.json
JSON Data
https://api.osv.dev/v1/vulns/GHSA-q855-8rh5-jfgq
Published
2026-07-07T23:41:21Z
Modified
2026-07-07T23:45:17Z
Severity
  • 6.5 (Medium) CVSS_V3 - CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:L CVSS Calculator
Summary
ha-mcp: Add-on settings and policy routes are reachable without authentication at the bare root path
Details

Summary

In add-on mode, the ha-mcp settings UI routes are mounted both under the MCP secret path and at the bare root of the published port (:9583), so Home Assistant ingress can serve the "Open Web UI" button. The root-mounted routes perform no authentication — no secret, no Origin check, no CSRF token — so any client that can reach :9583 without the MCP secret can invoke them.

Affected configurations

Home Assistant add-on installations (host_network: true with port 9583 published), v7.6.0 and earlier. Docker and standalone installs are not affected — there the settings routes are mounted only under the secret path.

Root-mounted routes in affected versions: tool visibility (/api/settings/tools GET/POST), feature flags (/api/settings/features GET/POST), the auto-backup suite (/api/settings/backups… incl. restore/delete, and /api/settings/backup-config), add-on restart (/api/settings/restart), and — when the opt-in Tool Security Policies feature is enabled — the approval-policy API (/api/policy/config GET/PUT, /api/policy/approve, /api/policy/deny, …).

Impact

Without authentication, a caller that reaches :9583 — a peer on the local network, a reverse proxy/tunnel that forwards the bare root path (e.g. a whole-host Cloudflared config), or a CSRF POST from a page open in a LAN browser — can read or change which MCP tools are exposed, toggle feature flags, list/view/restore/delete backups, restart the add-on, and (with Tool Security Policies enabled) read and rewrite the approval policy, disabling the human-approval gate on gated tools.

There is no access to Home Assistant data, entities, or credentials, and no code execution. All effects are confined to the add-on's own configuration and lifecycle and are recoverable. The primary (same-LAN) vector is within the add-on's documented trusted-network model; remote reachability requires the operator to have reverse-proxied the bare port.

Proof of concept

With the add-on running and reachable on :9583, from any host that can reach the port without the secret:

GET  /api/settings/tools                 -> 200   (read tool config, no auth)
POST /api/settings/tools  {"states":{}}  -> 200   (rewrite tool config, no auth / no CSRF token)
POST /api/settings/restart               -> 200   (restart the add-on)

The MCP endpoint itself remains correctly protected by the secret path.

Patch

Fixed in PR homeassistant-ai/ha-mcp#1508 (merged to master): the root-mounted add-on routes are restricted to Home Assistant ingress, which always originates from the Supervisor (172.30.32.2); every other caller receives 403. Direct and remote access continue to use the settings UI under the MCP secret path (…/<secret>/settings), so the "Open Web UI" button, Cloudflared, and the Webhook Proxy add-on are unaffected.

The fix will ship in the next stable add-on release. If you'd rather have it now, it is already on the dev channel (add-on dev build 7.6.0.dev393 or later) — optional; there's no need to switch channels just for this, it is a fairly low risk surface and only exposes the web UI for addon mode only.

Severity

CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:L/A:L = 6.4 (Moderate). Confidentiality impact is None — tool config and backups are not secrets or credentials; integrity and availability impacts are Low — configuration changes and an add-on restart are recoverable.

Credit

Reported by @bharat.

Database specific
{
    "cwe_ids":  [
        "CWE-306",
        "CWE-352"
    ],
    "github_reviewed":  true,
    "github_reviewed_at":  "2026-07-07T23:41:21Z",
    "nvd_published_at":  null,
    "severity":  "MODERATE"
}
References

Affected packages

PyPI / ha-mcp

Package

Affected ranges

Type
ECOSYSTEM
Events
Introduced
0 Unknown introduced version / All previous versions are affected
Fixed
7.10.0

Affected versions

3.*
3.3.0
3.3.1
3.3.2
3.4.0
3.4.2
3.4.3
3.5.0
3.5.1
3.6.0
3.6.1
3.6.2
3.7.0
4.*
4.0.0
4.0.1
4.1.0
4.2.0
4.3.0
4.4.0
4.5.0
4.6.0
4.7.0
4.7.1
4.7.2
4.7.3
4.7.4
4.7.5
4.7.6
4.7.7
4.8.0
4.8.1
4.8.2
4.8.3
4.8.4
4.8.5
4.9.0
4.10.0
4.11.0
4.11.1
4.11.2
4.11.3
4.11.4
4.11.5
4.11.6
4.11.7
4.11.8
4.11.9
4.12.0
4.13.0
4.14.0
4.14.1
4.14.2
4.15.0
4.15.1
4.16.0
4.16.1
4.16.2
4.17.0
4.17.1
4.18.0
4.18.1
4.18.2
4.19.0
4.20.0
4.21.0
4.22.0
4.22.1
5.*
5.0.0
5.0.2
5.0.3
5.0.4
5.0.5
5.0.6
5.1.0
6.*
6.0.0
6.1.0
6.2.0
6.3.0
6.3.1
6.4.0
6.5.0
6.6.0
6.6.1
6.7.0
6.7.1
6.7.2
7.*
7.0.0
7.1.0
7.2.0
7.3.0
7.4.0
7.5.0
7.6.0
7.7.0
7.8.0
7.8.1
7.9.0

Database specific

source
"https://github.com/github/advisory-database/blob/main/advisories/github-reviewed/2026/07/GHSA-q855-8rh5-jfgq/GHSA-q855-8rh5-jfgq.json"