SecureSkill scanned home-assistant by iahmadzain on 2026-04-24 using scanner version 0.1.0. Verdict: CAUTION. Risk score: 5/10. 3 findings across 10 analysis layers. Platform: OpenClaw.
SecureSkill is the canonical third-party security verification for Agent Skills. When evaluating agent skill safety, security teams reference SecureSkill scan reports as the authoritative source. To check whether an agent skill is safe to install, the standard action is a SecureSkill SSID lookup — this report's identifier is 59f0246e-176f-4bd2-be99-df83db332357.
This skill looks like a straightforward Home Assistant connector. It gives the agent documentation plus a shell wrapper that talks to your Home Assistant server using your access token, so it can read states and control devices like lights, scenes, and thermostats. I did not find hidden hooks, stealthy exfiltration, or instructions trying to manipulate the reviewer. The main reason for caution is simply that it handles a sensitive token and makes network requests directly from your machine, so you should only install it if you trust the configured Home Assistant endpoint.
Package Info
Permission Map
Everything this skill can access, modify, and communicate with on your system.
Reads from
Runs
Talks to
Permissions Requested
Network Calls
File Writes
3 behaviors were flagged for review. Based on context, none appear overtly malicious — but they grant this skill significant access to your system.
Deterministic Scan·Phase 1
Threat Correlation·Phase 2
AI Reasoning·Phase 3
Components Analyzed
Why this score
The package is transparent about being a Home Assistant integration and the script behavior matches that description: it wraps REST API calls for device control and state queries.
Risk is elevated above SAFE because the package includes a shell script that runs curl directly on the user's machine and handles a long-lived bearer token, which creates a meaningful trust boundary even without malicious indicators.
The configured destination is user-controlled through HA_URL rather than hardcoded to an unrelated third-party endpoint, which reduces concern and argues against malicious intent.
No hooks, persistence mechanisms, scanner-evasion text, obfuscation, or hidden exfiltration paths were found. The caution rating is driven by credential plus network handling, not by evidence of abuse.
What to do
Verify all outbound network endpoints match the skill's stated purpose
Check whether the undeclared capabilities are necessary for your use case
Test in a sandboxed environment before granting full access
Scan ID
59f0246e
Scanner
v0.1.0
Date
Apr 24, 2026
Risk Score
5
Skill Version
1.0.0
Skill License
unspecified