The Agent Skills Directory

[SAFE]: The skill consists of technical documentation and code recipes for the @commercengine/seo library. Analysis of the instructions and provided code snippets found no malicious patterns, prompt injections, or unauthorized data access attempts.
[EXTERNAL_DOWNLOADS]: The skill references the official vendor package @commercengine/seo and documentation at llm-docs.commercengine.io. These are trusted vendor resources associated with the author 'commercengine'.
[INDIRECT_PROMPT_INJECTION]: The skill provides instructions for ingesting and processing catalog data from the Commerce Engine API to generate metadata and markdown mirrors.
Ingestion points: Catalog data retrieved via API calls such as getProductDetail and listCategories (described in references/head-and-jsonld.md).
Boundary markers: The documentation suggests using conditional logic in framework layouts to manage sitewide vs. page-specific metadata, reducing the risk of accidental instruction confusion.
Capability inventory: The skill provides logic for rendering metadata and markdown content; it does not expose dangerous capabilities like arbitrary command execution or system-level file writes based on the processed data.
Sanitization: The documentation explicitly mentions that serializeJsonLd is used to escape characters (e.g., <, >, &) in JSON-LD output, a key security best practice for handling potentially untrusted data.

ce-seo