The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill implements logic to process untrusted data from external sources.
Ingestion points: Webhook request bodies in references/webhook-observability.md and API response extensions in references/instrumented-graphql-client.md.
Boundary markers: The code includes HMAC signature verification for webhooks to validate the source.
Capability inventory: The skill is granted Read, Write, and Edit tool permissions.
Sanitization: The implementation in references/structured-logging.md includes automatic PII redaction for logged Shopify request context.

shopify-observability