The Agent Skills Directory

[SAFE]: Detailed analysis of the skill instructions and the referenced implementation playbook confirms that the content is exclusively educational reference material. The skill focuses on privacy-by-design principles, consent management, and breach notification without exhibiting malicious patterns, obfuscation, or unauthorized data exfiltration.
[INDIRECT_PROMPT_INJECTION]: The implementation patterns in resources/implementation-playbook.md (specifically the DSARHandler and DataMinimization examples) describe systems that ingest and process external user data.
Ingestion points: Data is ingested through request detail objects in the submit_request method and form data in the collect_only_needed method.
Boundary markers: The skill does not define specific prompt boundary markers, as these are provided as implementation templates for developers rather than direct agent instructions.
Capability inventory: The patterns utilize database writes, event bus notifications, and local file reads for IP-to-country geolocation mapping.
Sanitization: The skill explicitly recommends and provides code for sanitization and mitigation, including PII hashing, data minimization through field filtering, and the use of rotating pseudonyms.

gdpr-data-handling