The Agent Skills Directory

[SAFE]: The skill implementation focuses on product feed optimization, custom label strategy, and competitive pricing analysis. All provided code snippets are functional templates for data processing and auditing.- [INDIRECT_PROMPT_INJECTION]: The skill ingests external product data which could potentially contain malicious instructions.
Ingestion points: The auditProductFeed function in SKILL.md consumes data from getMerchantProducts and getFeedDiagnostics.
Boundary markers: The instructions do not define specific boundary markers for the processed product attributes.
Capability inventory: The skill performs data analysis and returns an audit report; no high-risk capabilities such as command execution, arbitrary file writes, or network exfiltration were identified.
Sanitization: No input validation or sanitization logic is applied to the product titles, descriptions, or other attributes before they are processed by the audit functions.

shopping-ads