vardoger — Analyze History

• Outcome: Review
• Latest audit: May 29, 2026 at 9:35 PM UTC
• Version:
  0.3.1

Overview

This skill is transparent about its goal, but it reads broad past conversation history and writes persistent global assistant personalization, so users should review it carefully before installing.

Install only if you want an assistant to inspect past OpenClaw conversations and turn them into persistent global instructions. Verify the vardoger CLI source and version, approve only the expected vardoger commands, review the generated markdown before keeping it, and know where to remove the generated personalization file if it captures anything sensitive or unwanted.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• 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

Findings (4)

Vague Triggers

Medium

• Confidence: 86% confidence
• Finding: The activation condition is overly broad because merely mentioning "vardoger" can trigger a workflow that reads historical conversations and writes persistent personalization files. This increases the chance of unintended execution of a high-sensitivity skill without clear user intent or informed consent.

Ssd 3

High

• Confidence: 96% confidence
• Finding: The skill is explicitly designed to read the user's full conversation history from platform session directories and derive behavioral profiles from it. Even if intended as personalization, this creates a powerful data-exposure pathway because sensitive prior content may be processed, summarized, and reused beyond its original context.

Ssd 3

High

• Confidence: 95% confidence
• Finding: The skill directs the assistant to read batch outputs containing raw conversation data and produce summaries of behavioral signals. Summarization does not eliminate risk: sensitive data can be retained, inferred, or transformed into derived artifacts that persist or are shown to the user.

Ssd 3

High

• Confidence: 97% confidence
• Finding: Writing synthesized personalization into a persistent global rules file extends the lifetime and scope of information derived from past conversations. This can cause sensitive preferences, private facts, or inferred traits to be silently reused across future sessions and contexts, amplifying privacy and data-retention risk.

VirusTotal

63/63 vendors flagged this skill as clean.