creating-glaido-mcp-servers
Fail
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The documentation and validator script provide instructions to install the 'uv' package manager via 'curl -LsSf https://astral.sh/uv/install.sh | sh', which executes a remote script from the official Astral domain.
- [REMOTE_CODE_EXECUTION]: Instructions in SKILL.md use 'git clone' to fetch example server code from the author's GitHub repository (github.com/daveebbelaar/glaido-skills.git).
- [COMMAND_EXECUTION]: The utility script 'scripts/validate_glaido_mcp.py' uses 'subprocess.Popen' to launch and verify user-defined MCP server commands as part of the validation process.
- [DYNAMIC_EXECUTION]: The validation tool performs runtime execution of the generated server code to ensure protocol compliance and stability.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of MCP servers that serve as tool providers, creating a surface for indirect prompt injection via tool outputs.
- Ingestion points: Data returned by tools in the scaffolded Python and TypeScript servers.
- Boundary markers: No explicit boundary markers are provided in the tool response templates.
- Capability inventory: The validator script possesses 'subprocess' capabilities, and generated servers are intended to perform network and file operations.
- Sanitization: Templates include Zod (TypeScript) and FastMCP type hints (Python) to validate and sanitize agent-provided inputs.
Recommendations
- HIGH: Downloads and executes remote code from: https://astral.sh/uv/install.sh - DO NOT USE without thorough review
Audit Metadata