The Agent Skills Directory

[COMMAND_EXECUTION]: Path Traversal Vulnerability
Evidence: Found in scripts/design_system.py within the persist_design_system function.
Details: The script constructs file paths using the project_name and page arguments by simply replacing spaces with hyphens. It does not sanitize for directory traversal sequences such as ../. If an AI agent is manipulated into passing a malicious string to these parameters, it could result in the agent writing or overwriting arbitrary files on the local system outside of the intended directory.
[EXTERNAL_DOWNLOADS]: Recommendation of Unverifiable CLI Tool
Evidence: Found in references/upstream-README.md.
Details: The documentation encourages the global installation of uipro-cli via npm (npm install -g uipro-cli). This package is maintained by an unverified third-party account (nextlevelbuilder) and does not originate from a recognized trusted organization, posing a supply chain risk to users who follow the installation instructions.
[INDIRECT_PROMPT_INJECTION]: Data Ingestion and Capability Surface
Ingestion points: scripts/core.py and scripts/design_system.py read data from numerous bundled CSV files in assets/data/.
Boundary markers: While data is structured in CSV format, there are no explicit instructions for the agent to disregard natural language instructions that might be embedded within the CSV data fields.
Capability inventory: The skill possesses file-writing capabilities through the persist flag, which generates and saves markdown files to the local disk.
Sanitization: There is no evidence of sanitization or validation of the CSV content before it is interpolated into prompts or written to files.

ui-ux-pro-max