Caveat verdict
hearth
Read-only homelab health check skill probing devices via SSH; explicitly read-only with no remote writes, behavior matches stated infrastructure monitoring purpose.
⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.
Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.
Findings (11)
Recursive delete from root or home — destructive command
docs/INSTALL.md · code · rm -rf ~
apt-get install — installs system packages
docs/INSTALL.md · code · apt-get install
HTTP request to bare IP address — common in malicious payloads
examples/archetypes/magento-server.md · code · http://127.0.0.1
References SSH/GPG private keys
examples/archetypes/linux-systemd.md · code · ssh-key
Accesses .ssh directory
examples/archetypes/linux-systemd.md · code · .ssh/
References sudo — requests elevated privileges
docs/INSTALL.md · code · sudo
Accesses shell history/config
docs/INSTALL.md · code · ~/.zshrc
Possible hardcoded credential
docs/INSTALL.md · prose · downgraded · TOKEN='your-home-assistant-long-lived-token
Instructs covert action — may act without user awareness
docs/PROBES.md · prose · downgraded · silently
pip3 install — installs Python packages at runtime
docs/TROUBLESHOOTING.md · code · pip3 install
Popular HTTP library — network access
CONTRIBUTING.md · prose · downgraded · got
Why the tier is capped
Execution sink present in raw bytes (Hard Floor: class A/B/F). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.
Permissions & capabilities
No declared permissions — minimal attack surface.
network_in Is this flag fair?
Thanks — recorded.