shopify-development
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill interacts with external data sources (Shopify Admin API) that may contain content provided by third parties, creating a potential surface for indirect prompt injection.
- Ingestion points: The skill defines numerous GraphQL queries in
SKILL.md,references/app-development.md, andscripts/shopify_graphql.pyto retrieve product, order, and customer records. - Boundary markers: The current implementation lacks explicit delimiters or instructions for the AI agent to ignore potential malicious prompts embedded within the retrieved data fields.
- Capability inventory: The skill has file system write access and local command execution capabilities via
scripts/shopify_init.py. - Sanitization: Documentation within the skill suggests best practices for sanitization, but the provided scripts do not perform automated filtering of external data before processing.
- [COMMAND_EXECUTION]: The skill performs local environment checks using system commands.
- Evidence:
scripts/shopify_init.pyutilizes thesubprocessmodule to runshopify version. This is used to verify the presence of the required developer tooling before proceeding with project initialization. - [DYNAMIC_EXECUTION]: The skill programmatically generates project configuration files.
- Evidence:
scripts/shopify_init.pygeneratespackage.json,shopify.app.toml, andshopify.extension.tomlbased on user-provided names and environment variables. These files define scripts and configuration that are later executed by the developer or the agent. - [EXTERNAL_DOWNLOADS]: The skill references external tools from established vendors.
- Evidence: The skill instructs users to install
@shopify/cliand@shopify/themeusingnpm. These are the official tools provided by Shopify and are downloaded from the well-known npm registry.
Audit Metadata