Back to skill

Security audit

NoteGPT Browser Automation Skill

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this browser automation skill benign with medium confidence: its reported behavior broadly matches its purpose, though the scanner flags unpinned installation and browser updates as supply-chain risks.

The verdict is benign, not risk-free. Before installing, consider whether you trust the package and browser update sources, and review or pin versions where possible. Treat form submissions as consequential and check them before proceeding. Skillspector could not resolve two local path references, so some evidence is incomplete.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

MCP server referenced without pinned version: 'npx cloakbrowser'.

Medium
Category
MCP Rug Pull
Confidence
88% confidence
Finding

The MCP classification is incorrect, but npx cloakbrowser install can resolve and execute an unpinned package when a local installation is unavailable. The skill additionally describes downloading a browser executable and enabling background updates, creating a supply-chain exposure in which a compromised upstream release could execute with the agent's permissions; no compromise is established by this document.

Content

No source excerpt is available for this finding.

The form-automation example clicks Submit without warning that it may send data or commit changes on the destination site.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The executable example instructs the agent to click '#submit', which can submit a form and cause server-side changes. The skill provides no warning about these effects or requirement to review the form and obtain authorization before submission.

Content

No source excerpt is available for this finding.