skills/involvex/skills/mcp-cli/Gen Agent Trust Hub

mcp-cli

Pass

Audited by Gen Agent Trust Hub on Apr 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill acts as a documentation layer for the mcp-cli utility. Analysis of the instructions and examples revealed no malicious intent, prompt injection, or hidden payloads.
  • [COMMAND_EXECUTION]: The skill facilitates the execution of mcp-cli to interact with external tools. This is the intended functionality of an MCP client.
  • [DATA_EXPOSURE]: Examples demonstrate reading local filesystem data (e.g., read_file). This behavior is consistent with the primary purpose of MCP-integrated tools and is documented clearly.
  • [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by ingesting data from external MCP servers.
  • Ingestion points: Output from mcp-cli (SKILL.md).
  • Boundary markers: None explicitly defined in the workflow examples.
  • Capability inventory: Execution of mcp-cli which can trigger filesystem, database, or API operations.
  • Sanitization: No validation or sanitization of tool outputs is mentioned; however, this is a standard characteristic of MCP client documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 8, 2026, 09:27 PM