ClawHub
Back to skill

Security audit

Academic Results Writer

Security checks across malware telemetry and agentic risk

Overview

This is a coherent academic writing skill; its main caveat is a disclosed automatic file-output mode that writes long results drafts to a Desktop folder.

Before installing, be aware that long generated Results drafts may be saved automatically under ~/Desktop/OpenClaw_Paper_Analysis/outputs_md/results_writer/. Avoid feeding confidential or unpublished research data unless you are comfortable with that local file being created, and check for filename collisions if re-running similar analyses.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Description-Behavior Mismatch

Low
Confidence
90% confidence
Finding
The skill instructs the agent to write output to a fixed path on the user's Desktop, which introduces an unnecessary side effect beyond text generation. Even though the content is academic writing rather than code execution, fixed-path file writes can overwrite existing files, expose sensitive research content in a predictable location, and violate least-privilege expectations for a writing-only skill.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The example explicitly instructs the agent to write output to a specific local filesystem path, but does not disclose this as a side effect or require explicit user consent. In an agent setting, undocumented file writes can surprise users, overwrite existing files, or be combined with path assumptions to create unsafe local state changes.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.