Daily Review

Security checks across malware telemetry and agentic risk

Overview

The skill's high-level purpose (aggregating personal productivity signals) matches most of the code, but there are multiple incoherences and risky practices — notably hardcoded credentials and a hardwired SSH target — that make this unsafe to install without changes.

This skill aggregates lots of personal data and in its current form is risky to install: 1) It expects many secrets (Google Workspace, Slack, Fireflies, Bird/X cookies) but the registry doesn't declare them — verify what secrets you'll need and store them securely. 2) The script contains hardcoded credentials (an X/Bird auth token and ct0 cookie) and a hardcoded SSH user/host; do NOT run this as-is — those hardcoded values expose credentials and indicate the package may be tailored to a specific person's environment. 3) The script SSHes into a remote Mac and reads the Screen Time database and ActivityWatch on that host; confirm you control/trust that remote machine before allowing SSH or running the script. 4) Before installing: ask the publisher for a version that removes embedded secrets, makes remote host and usernames configurable, and documents exactly which tokens/files are required; review/replace hardcoded tokens; audit any third-party code you will git-clone (bird CLI). 5) If you still want to use it, run it in a sandbox or isolated account, rotate any exposed tokens, and only provide the minimum credentials necessary (prefer short-lived or scoped tokens). Given the hardcoded credentials and hardwired SSH target, treat this package as untrusted until those issues are fixed.

SkillSpector

By NVIDIA

SkillSpector findings are pending for this release.

VirusTotal

No VirusTotal findings

View on VirusTotal