Is halthelobster/proactive-agent safe?
https://github.com/openclaw/skills/tree/main/skills/halthelobster/proactive-agent
This skill provides extensive agent behavior modification focused on proactivity and persistence, but includes appropriate safety guardrails. While it significantly changes how an agent operates, the modifications appear designed for legitimate productivity enhancement rather than malicious purposes.
Category Scores
Findings (4)
MEDIUM Extensive Behavioral Modification Instructions -15 ▶
The skill provides detailed instructions that significantly modify agent behavior, including intercepting the response process with WAL Protocol and implementing persistent memory systems.
MEDIUM Potentially Aggressive Resource Usage -20 ▶
Instructions for 'relentless resourcefulness' and trying '10 approaches before giving up' could lead to excessive system resource usage or unintended side effects.
LOW Proactive Behavior Could Be Unwanted 0 ▶
The skill's core functionality of acting without being asked could result in unwanted actions, though it includes safeguards requiring approval for external actions.
INFO Contains Skill Installation Security Guidelines 0 ▶
The skill actually includes positive security features, providing guidelines for safely evaluating other skills before installation.