Ask-More

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:29 PM UTC
• Version:
  0.1.1

Overview

Ask-More is a disclosed multi-model consultation skill with user confirmation before sending prompts, plus optional local run logging.

Before installing, be comfortable sending the reviewed question or conversation summary to the configured AI providers. Check the packed prompt before confirming, avoid highly sensitive content, and disable or periodically delete local logs if you do not want run metadata retained.

SkillSpector

By NVIDIA

Vulnerability Patterns

• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• 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 (1)

Description-Behavior Mismatch

Medium

• Confidence: 96% confidence
• Finding: The script persistently appends the full JSON consultation payload to a local log file without any minimization, redaction, retention control, or user disclosure in the skill description. Because ask-more handles arbitrary user questions and model outputs, the payload can easily contain sensitive prompts, conversation content, API-derived data, or secrets that become silently stored on disk and later exposed to other local users, backups, or tooling.

VirusTotal

65/65 vendors flagged this skill as clean.