Skills
skills/dkyazzentwatwa/chatgpt-skills/mcp-builder/Gen Agent Trust Hub

mcp-builder

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • COMMAND_EXECUTION (MEDIUM): The MCPConnectionStdio class in scripts/connections.py uses mcp.client.stdio.stdio_client to spawn subprocesses based on provided command and argument parameters. While this is the standard mechanism for MCP stdio transport, it allows for arbitrary command execution if the connection parameters are influenced by untrusted external data.
  • EXTERNAL_DOWNLOADS (LOW): The MCPConnectionSSE and MCPConnectionHTTP classes in scripts/connections.py facilitate network requests to external URLs. This functionality represents a surface for potential Server-Side Request Forgery (SSRF) or data exfiltration if the agent is directed to connect to malicious endpoints.
  • INDIRECT_PROMPT_INJECTION (LOW): The call_tool method in scripts/connections.py retrieves content from external MCP servers. This content is an ingestion point for untrusted data that could contain hidden instructions if subsequently processed by an LLM without proper sanitization or boundary markers.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:04 PM