paddle-catalog-setup
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-defined product names, descriptions, and pricing details which are interpolated into executable code blocks. This creates a potential vulnerability where malicious input could influence code execution.
- Ingestion points: User input regarding product and price configurations in SKILL.md.
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are included in the generated code templates.
- Capability inventory: The skill utilizes powerful tools including the
executetool for JavaScript execution and local file-writing for script generation. - Sanitization: There is no evidence of sanitizing or escaping user input before it is used in code generation.
- [DYNAMIC_EXECUTION]: The skill's operational logic depends on the runtime generation and execution of code strings to interact with the Paddle API via MCP tools and generated Node.js scripts.
- [EXTERNAL_DOWNLOADS]: The skill recommends installing the official
@paddle/paddle-node-sdkpackage from the npm registry to support catalog seeding. This is an expected operation for the skill's functionality.
Audit Metadata