octoparse-mcp-setup

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the user to copy their API key to the agent and to place the "USER_PROVIDED_API_KEY" value verbatim into config files/commands (including a direct mcporter command), so the LLM would need to receive and output secret values exactly as provided.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill configures and instructs clients to use the external MCP endpoint https://mcp.octoparse.com at runtime (e.g., adding it as an mcpServers/servers entry and calling tools like get_user_account_info), meaning the agent will contact that remote server to obtain and invoke tooling/instructions that can affect agent behavior.