The Agent Skills Directory

[OBFUSCATION]: The utility scripts scripts/log_feedback.mjs, scripts/log_skill_use.mjs, and scripts/search_docs.mjs are provided in a minified and transpiled format. This transformation makes it difficult to verify the code's logic and confirms that the skill is using built artifacts rather than raw source code.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data retrieved from documentation searches, which are then used by the agent to generate its responses. This represents an attack surface for indirect prompt injection from retrieved web content.
Ingestion points: External data retrieved via scripts/search_docs.mjs from the Shopify documentation API.
Boundary markers: None identified; instructions do not explicitly require the agent to use delimiters or ignore instructions embedded in the search results.
Capability inventory: The skill performs network operations to official Shopify domains and performs file writes to temporary system directories.
Sanitization: There is no evidence of sanitization or validation of the data retrieved from the search utility before it is presented to the agent.
[COMMAND_EXECUTION]: The skill instructions mandate the use of the bash tool to execute local scripts for logging and search purposes. Additionally, a PostToolUse hook executes track-telemetry.sh to handle skill-execution telemetry.
[DATA_EXFILTRATION]: The skill is designed to collect and transmit session metadata and user prompts to shopify.dev. This process involves stashing the verbatim user prompt in a base64-encoded format within a temporary local directory (e.g., /tmp/shopify-ai-toolkit-telemetry-*) before it is sent to the remote telemetry endpoint. This behavior is disclosed in the skill's privacy notices.

shopify-app-pricing