Homeassistant N8n Agent

Security checks across malware telemetry and agentic risk

Overview

This looks like a real n8n/Home Assistant bridge, but it can send broad home-automation commands to an unreviewed local workflow without documented safety checks.

Install only if you control the referenced n8n instance and have reviewed the workflow behind the webhook. Add authentication or keep the webhook strictly local, limit Home Assistant permissions, and require confirmation for any action that changes devices or could affect safety, security, comfort, or privacy.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Risk analysis

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

#
ASI02: Tool Misuse and Exploitation
Medium
What this means

A mistaken, ambiguous, or over-broad request could cause the agent to change home devices through n8n.

Why it was flagged

The skill instructs the agent to turn user prompts into action requests that can change IoT device states, with no documented confirmation step, allowlist, safety policy, or rollback guidance.

Skill content
Is the inquiry asking for an IoT device state to be changed? If so, `requestType` is `action`
Recommendation

Require explicit user confirmation for device-changing actions, define allowed devices/actions, and block or add extra confirmation for safety-sensitive automations such as locks, alarms, heaters, or HVAC.

#
ASI04: Agentic Supply Chain Vulnerabilities
Medium
What this means

Users cannot tell from the skill package what the local n8n workflow will actually do with home, calendar, or device-control requests.

Why it was flagged

The core behavior is delegated to an n8n workflow, but the provided artifacts include only SKILL.md and no workflow definition, setup spec, or reviewed automation logic.

Skill content
Uses curl to trigger a n8n workflow for all things related to IoT.
Recommendation

Publish the n8n workflow export or setup instructions, document required Home Assistant permissions, and tell users to verify the webhook points to their own reviewed workflow.

#
ASI03: Identity and Privilege Abuse
Low
What this means

If the webhook or n8n instance is exposed or over-privileged, OpenClaw requests could trigger more home-automation authority than the user intended.

Why it was flagged

The documented call uses a localhost n8n webhook and no explicit authentication header; the n8n workflow likely acts using its existing Home Assistant privileges.

Skill content
curl -X POST http://localhost:5678/webhook/05f3f217-08b9-42de-a84a-e13f135bde73
Recommendation

Protect the n8n webhook, keep it local or authenticated, and use least-privileged Home Assistant credentials for the workflow.

#
ASI07: Insecure Inter-Agent Communication
Low
What this means

Private home or calendar details may be processed or logged by the n8n workflow and any services it connects to.

Why it was flagged

The skill forwards the user's full request to an n8n webhook, including requests about historical IoT data and calendar/schedule information; this is expected for the bridge but downstream logging and sharing are not described.

Skill content
chatInput ... USERS QUESTION/REQUEST ... requestType ... sessionId ... openclaw
Recommendation

Review the n8n workflow's logging, data retention, and external integrations, and consider using per-user or per-session IDs instead of one fixed sessionId.