travel-requirements-expert
Pass
Audited by Gen Agent Trust Hub on Feb 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill utilizes a provided script
scripts/create_requirements_folder.pyto create a timestamped directory structure. The script is limited to standard filesystem operations using thepathlibandoslibraries and implements length-limiting and character-filtering (slugify) on folder names. - [EXTERNAL_DOWNLOADS] (SAFE): Uses Perplexity and Exa MCP servers for web-based travel research. These tools are used as intended for the skill's primary function and do not download executable code.
- [PROMPT_INJECTION] (LOW): Indirect Prompt Injection Surface: The skill ingests raw user input and writes it directly into
00-initial-request.mdwithout boundary markers or sanitization of the content. Evidence Chain: 1. Ingestion points:user_requestargument increate_requirements_folder.py. 2. Boundary markers: Absent in generated markdown files. 3. Capability inventory: Local file modification and network access via MCP. 4. Sanitization: Only the folder name is sanitized; the actual file content is written as raw text, which could be exploited if the agent re-processes these files as instructions.
Audit Metadata