shopify-liquid
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The package.json file defines dependencies on official Shopify packages such as @shopify/theme-check-node and @shopify/theme-check-common for Liquid linting and validation.\n- [DATA_EXFILTRATION]: The scripts scripts/search_docs.mjs and scripts/validate.mjs perform network requests to https://shopify.dev/ to facilitate documentation search and report validation instrumentation. This communication is directed to the official infrastructure of the skill's author, Shopify.\n- [COMMAND_EXECUTION]: The SKILL.md instructions require the agent to use the bash tool to run local scripts for component search and code validation. These operations use well-defined local scripts provided within the skill environment.\n- [PROMPT_INJECTION]: The skill defines a workflow vulnerable to indirect prompt injection where user-supplied requests influence generated Liquid code that is subsequently passed to a shell command. This could allow a malicious user to trigger shell command execution in the agent's environment.\n
- Ingestion points: User prompts processed during the component generation task in SKILL.md.\n
- Boundary markers: None used in the shell command construction in SKILL.md.\n
- Capability inventory: bash tool used to execute scripts/validate.mjs via a shell command.\n
- Sanitization: No escaping or validation is performed on the generated code before it is passed as an argument to the shell command.
Audit Metadata