ClawHub
Back to skill

Security audit

stock-market-dashboard

Security checks for vulnerabilities and agentic risk

Overview

This skill is a scoped market-dashboard generator that fetches read-only market data and creates a local static HTML report, though its metadata should more clearly acknowledge the file write.

Before installing, understand that this skill is not fully read-only in the filesystem sense: it fetches read-only market data using your SentiSense API key and writes a local static HTML dashboard. Do not treat the generated report as live market data or investment advice, and avoid embedding secrets in the output.

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
  • 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

Medium
Confidence
99% confidence
Finding
The skill metadata claims 'No write operations' and 'Read-only,' but the body explicitly instructs the agent to create and save an HTML file. This mismatch can mislead users, reviewers, or policy engines about the skill's actual side effects, causing file writes to occur in contexts where only read-only behavior was expected.

Intent-Code Divergence

Medium
Confidence
98% confidence
Finding
The skill repeatedly presents itself as read-only and non-writing, yet later directs the agent to 'write the HTML yourself' and explicitly grants file-writing permission. This is a documentation/permission inconsistency that can undermine trust boundaries and automated safety decisions, especially in environments that distinguish read-only from filesystem-mutating skills.

Static analysis

No suspicious patterns detected.