Openclaw Proactive Agent Lite

• Outcome: Pass
• Latest audit: May 29, 2026, 8:33 PM
• Version:
  1.0.0

Overview

This text-only skill encourages more proactive agent behavior, but it does not install code, request credentials, or grant external access.

Install only if you want the agent to make more unsolicited suggestions and maintain more continuity across sessions. Use it first in lower-risk workflows, monitor what the host agent remembers or suggests, and avoid pairing it with credentials or high-impact tools unless those integrations have separate, clear approval controls.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

Medium

• Confidence: 81% confidence
• Finding: The README states that the agent will 'automatically begin exhibiting proactive behavior' with no additional configuration, but it does not define explicit triggers, scope limits, or approval boundaries. In a skill focused on memory, reverse prompting, and self-healing, this broad activation language can encourage unsolicited actions or overreach by downstream agents, increasing the risk of unintended behavior.

Vague Triggers

Medium

• Confidence: 89% confidence
• Finding: The skill claims it enhances 'any OpenClaw agent' and will automatically enable proactive behavior, but it does not define scope limits, activation conditions, or guardrails. That ambiguity can cause the skill to influence agents and tasks beyond the user's expectations, increasing the risk of unintended autonomous behavior or policy drift when combined with other skills.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill advertises memory retention, reverse prompting, and self-healing behavior without disclosing how these features may affect output autonomy, persistence of context, or user control. In practice, this can mislead deployers into enabling behaviors that retain or reinterpret context in ways they did not intend, especially in environments handling sensitive data or requiring predictable outputs.

VirusTotal

63/63 vendors flagged this skill as clean.