building-python-mcp-servers
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is purely instructional and provides defensive programming guidance for MCP development.
- [COMMAND_EXECUTION]: The skill discusses using subprocess.run safely by avoiding shell=True and using shlex.split() for argument parsing.
- [PROMPT_INJECTION]: The content includes explicit warnings about treating tool inputs as untrusted data and includes a section on prompt-injection awareness.
- [SAFE]: Regarding Indirect Prompt Injection vulnerability surface: 1. Ingestion points: Tool function arguments in SKILL.md. 2. Boundary markers: The skill advises explicit validation checks. 3. Capability inventory: Use of subprocess.run and file reading operations. 4. Sanitization: Code examples show absolute path validation and existence checks.
Audit Metadata