Back to skill

Security audit

stock-market-dashboard

Security checks for vulnerabilities and agentic risk

Overview

The skill is marked suspicious because its instructions reportedly go beyond creating a named HTML file and call for persistent dashboard changes; scanner coverage is incomplete.

Review the dashboard persistence and replacement instructions before installing. Consider whether you want a skill to pin or replace a persistent dashboard, and note that scanning was partial because several local references could not be matched to bundled files; the results do not establish that all behavior is safe.

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
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 (1)

The delivery instructions prescribe persistent session-dashboard writes beyond the declared permission to write one user-named HTML file.

Medium
Category
Excessive Agency
Confidence
91% confidence
Finding

The skill declares permission to write one user-named HTML file, but its delivery instructions additionally prescribe pinning a widget that survives session resets and replacing content under a fixed dashboard name. Without explicit user approval, this expands a file-generation request into persistent application-state modification and could overwrite an existing widget. The static, sandboxed content reduces execution and exfiltration risks, but does not authorize persistence or replacement.

Content

No source excerpt is available for this finding.