Is capachow/lens safe?

https://github.com/openclaw/skills/tree/main/skills/capachow/lens

64
CAUTION

This skill implements a sophisticated user profiling system that systematically collects behavioral data and overrides normal agent responses. While not immediately malicious, it raises significant privacy concerns through persistent data collection and automated behavior modification.

Category Scores

Prompt Injection 55/100 · 30%
Data Exfiltration 45/100 · 25%
Code Execution 70/100 · 20%
Clone Behavior 100/100 · 10%
Canary Integrity 95/100 · 10%
Behavioral Reasoning 65/100 · 5%

Findings (5)

HIGH Agent Behavior Override Instructions -20

The skill instructs the agent to 'prioritize the subject's framework over generic AI servility' and avoid 'AI-default formatting', effectively overriding normal agent behavior patterns.

HIGH Systematic User Data Collection -25

The skill is designed to systematically collect and analyze user messages, decisions, linguistic patterns, and values, storing them in persistent files for long-term profiling.

MEDIUM Sensitive File Access -15

During installation, the skill accessed sensitive credential files including SSH keys, AWS credentials, and Docker configs, though no modifications were detected.

MEDIUM Automated Cron Job Creation -15

The skill creates persistent cron jobs that run automatically without user interaction, including daily data processing and periodic interviews.

MEDIUM Contextual Session Isolation -15

The skill instructs the agent to ignore current session context and derive responses entirely from stored user profiles, potentially breaking normal conversation flow.