ClawHub
Back to skill

Security audit

Skylight

Security checks across malware telemetry and agentic risk

Overview

This is a coherent unofficial Skylight integration, but it handles sensitive account credentials and tokens, so users should treat it carefully.

Install only if you are comfortable giving an unofficial, reverse-engineered Skylight integration access to your household calendar account. Keep SKYLIGHT_URL on the real Skylight domain, store passwords and bearer tokens in a secret manager where possible, do not commit or log tokens, avoid HTTPS proxy token capture unless you understand the certificate and traffic risks, and confirm before allowing the agent to create or change chores, events, lists, rewards, or task box items.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The README explicitly instructs users to capture authentication tokens using an HTTPS proxy, which encourages interception of highly sensitive session credentials without any warning about account takeover, token leakage, or handling of decrypted traffic. In the context of a reverse-engineered unofficial API, this is more dangerous because users may normalize unsafe credential collection practices and store reusable tokens that grant ongoing access to family/calendar data.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The README tells users to place email/password and authentication tokens in environment variables but provides no guidance on protecting those secrets from shell history, process inspection, logs, CI systems, or shared environments. Because this skill accesses a household calendar/rewards system tied to family data, exposed credentials could allow unauthorized access to personal schedules, chores, lists, and account functions.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The script directly submits a user's Skylight email and password to the remote service and exchanges them for a bearer token with broad 'everything' scope, but the file itself provides no warning, consent boundary, or guidance on secure handling of the resulting token. In an agent skill context, this is security-relevant because the script automates credential use and emits a reusable access token to stdout, increasing the chance of accidental leakage through logs, chaining, or misuse by downstream components.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.