ClawHub
Back to skill

Security audit

Research Assistant

Security checks for vulnerabilities and agentic risk

Overview

The skill is a local research-note organizer, but its export command can overwrite broadly chosen files in the user's home directory or /tmp, which is more authority than the purpose requires.

Review this skill before installing if you expect agents to handle untrusted prompts. The note database behavior is expected, but the export feature should ideally be limited to a dedicated export folder and should avoid overwriting existing files without explicit confirmation.

A.I.G

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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
Findings (1)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/research_organizer.py:119
Finding

Overly Broad and Bypassable Export Path Validation

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • 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 (1)

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 58)May include surrounding context.

For content planning

bash
# Plan content topics
research_organizer.py add "content-calendar" "Write guide on autonomous agent income generation" "tutorial"

# Store references
research_organizer.py add "content-calendar" "Reference: ClawHub marketplace at clawhub.com" "resource"

Static analysis

No suspicious patterns detected.