ClawHub
Back to skill

Security audit

Mb Text Workflow

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed markdown documentation workflow for maintaining local memory-bank files, with some privacy and approval-handling cautions but no evidence of hidden, destructive, or exfiltrating behavior.

Install this if you want an agent to maintain local memory-bank markdown records for a project. Before using it, confirm that memory-bank updates are intended, keep it scoped to the current project, and avoid recording secrets, credentials, personal data, or unnecessary verbatim private content in session, error, and edit-history files.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The document contains conflicting instructions: one section requires approval before creating a missing session file, while later workflow steps instruct creating the session file if it does not exist. In a skill that governs file updates, this inconsistency can cause an agent to perform unapproved writes despite the stated approval gate, weakening a key safety control.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The trigger list includes several broad natural-language phrases such as "update memory bank" and "update tasks.md", which can cause the skill to activate in situations broader than intended. In an agentic environment, unintended invocation can lead to unnecessary file reads/edits, accidental workflow execution, or confusion about whether the user intended documentation-only actions.

Ssd 3

Medium
Confidence
95% confidence
Finding
The workflow directs the agent to persist session context, task details, errors, change history, and user-provided work information across multiple memory-bank files. In a text-based agent skill, that creates a real data retention risk: sensitive prompts, secrets, internal paths, error traces, and user content may be copied into durable markdown records and later exposed to other sessions or operators.

VirusTotal

VirusTotal engine telemetry is currently stale for this artifact.

View on VirusTotal

Static analysis

No suspicious patterns detected.