ClawHub
Back to skill

Security audit

LaTeX Revision Tracker

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent LaTeX revision workflow skill with only expected, user-directed local document and version-control guidance.

Review generated shell commands before running them. In particular, run cleanup only from the intended manuscript directory after confirming the diff PDF is good, and treat git push/tag examples as actions that can affect a shared remote repository.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The README suggests activation phrases like "Help me integrate this new section into my paper" and "Set up version control for my thesis," which are broad, natural-language requests likely to occur in normal agent conversations. If the hosting agent uses loose skill routing or automatic invocation based on semantic similarity, the skill could trigger unintentionally and influence tasks beyond the user's explicit intent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill includes an rm -f cleanup command that deletes multiple files but does not clearly warn about irreversible deletion or advise users to verify paths and filenames first. In a workflow-oriented skill, users may copy-paste commands directly; if adapted carelessly or run from the wrong directory, this can cause unintended data loss, especially because .tex and auxiliary files may still be needed for debugging or regeneration.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.