ClawHub
Back to skill

Security audit

Research to Sheets

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed workflow for web research and user-confirmed Google Sheets writes through ClawLink, with no hidden execution or destructive behavior found.

Before installing, understand that this skill will ask you to install and allow the ClawLink plugin, connect Google Sheets through ClawLink, and let the agent write to spreadsheets only after confirmation. Use care when collecting personal data such as emails or LinkedIn profile details: collect only data you are allowed to use, keep the scope small, and review previews before approving writes.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill explicitly encourages collecting contact names, roles, and emails from websites and writing them into Google Sheets, but it provides no privacy, consent, data-minimization, or legal-use guidance. That omission increases the risk of bulk collection of personal data for lead generation or profiling without adequate safeguards, especially because the workflow emphasizes normalization and storage of contact data.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 142)May include surrounding context.

md
## Rules

- Always use ClawLink tools for Google Sheets writes. Do not ask the user for separate Sheets credentials.
- Do not invent Google Sheets tool names or schemas. Use the live ClawLink catalog in the current turn.
- Ask for confirmation before creating spreadsheets, appending rows, overwriting cells, or editing many records.
- Do not fabricate missing research data just to complete a table.

Static analysis

No suspicious patterns detected.