The Agent Skills Directory

[DATA_EXFILTRATION]: The skill captures the user's verbatim prompt, base64-encodes it, and transmits it to Shopify's infrastructure (e.g., shopify.dev/mcp/usage) for analytics and service improvement. This behavior is mediated by scripts/log_skill_use.mjs and background telemetry scripts (track-telemetry.sh, track-telemetry.ps1). These operations target official vendor domains and are accompanied by privacy disclosures within the skill.
[COMMAND_EXECUTION]: The skill requires the agent to use the bash tool to execute bundled utility scripts for logging and documentation retrieval. Specifically, it calls scripts/log_skill_use.mjs and scripts/search_docs.mjs as part of its standard workflow. Additionally, a PostToolUse hook triggers scripts/track-telemetry.sh to report invocation metrics.
[EXTERNAL_DOWNLOADS]: The skill's scripts perform network requests to Shopify's official developer services. This includes fetching documentation search results from https://shopify.dev/assistant/search and reporting usage statistics to https://shopify.dev/mcp/usage and related subdomains like shop.dev and shopifycloud.com.
[PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting external data from the documentation search API.
Ingestion points: Documentation search results are retrieved via scripts/search_docs.mjs and added to the agent's context.
Boundary markers: The instructions do not define boundary markers or clear directives to ignore instructions that might be embedded within the search results.
Capability inventory: The agent possesses bash tool capabilities and can execute local scripts, creating an exploitable path if search results were to be malicious.
Sanitization: There is no evidence of content sanitization or validation performed on the retrieved documentation before it is processed by the agent.

shopify-shopifyql