Context Guardian Pro

• Outcome: Review
• Latest audit: May 29, 2026 at 9:33 PM UTC
• Version:
  1.0.0

Overview

This skill is not clearly malicious, but it is a review item because it is designed to run automatically, write to long-term memory, and claims a backup was made even though the code only writes a placeholder summary.

Install only if you are comfortable with automatic context monitoring and memory writes. Ask the publisher to add explicit opt-in scheduling, a disable path, review-before-save behavior, retention/deletion controls, real summarization, and clear disclosure of any future QQ transmission before relying on it for context backup.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (3)

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill explicitly describes automatic summarization, persistence to long-term memory files, and outbound QQ notifications, but it does not mention user consent, data minimization, retention limits, or warnings about external transmission. In a conversation-monitoring context, this can cause sensitive chat content to be stored or sent off-channel unexpectedly, increasing privacy and data leakage risk.

Missing User Warnings

Medium

• Confidence: 82% confidence
• Finding: The script automatically writes data to a persistent file under the user's home directory without any user confirmation or visibility beyond local logging. In an agent/skill context, silent persistence can create privacy and data-governance risks because conversation-derived content may be retained longer than intended and outside the user's expectations.

Ssd 3

Medium

• Confidence: 96% confidence
• Finding: The description explicitly states the skill will summarize conversation history, write it to long-term memory, and send a QQ alert when context thresholds are reached. That creates a credible data leakage path because potentially sensitive user content may be condensed and persisted or transmitted to an external service without any mention of minimization, redaction, consent, or access controls.

VirusTotal

67/67 vendors flagged this skill as clean.