mcp-management
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands to manage MCP servers, including environment setup with 'mkdir' and 'ln', and running local management scripts via 'bunx'.
- [EXTERNAL_DOWNLOADS]: The skill directs the user to install the 'gemini-cli' package globally from a public registry to enable automated tool execution.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting metadata from external MCP servers. 1. Ingestion points: 'assets/tools.json' (populated from external server tool/prompt/resource lists). 2. Boundary markers: No explicit markers or 'ignore' instructions are used when the LLM analyzes the catalog. 3. Capability inventory: The skill can execute arbitrary MCP tools and shell commands ('gemini', 'bunx'). 4. Sanitization: No validation or sanitization of remote metadata is performed before LLM processing.
Audit Metadata