The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were identified in the skill. A thorough analysis across all 10 threat categories confirms the content is instructional and follows safety best practices.- [PROMPT_INJECTION]: The skill does not contain instructions that attempt to override agent safety protocols or bypass constraints. It explicitly includes positive constraints instructing the agent to refuse misleading content and ensure honesty.- [DATA_EXFILTRATION]: No patterns indicating unauthorized data access, sensitive file harvesting, or suspicious network transmissions were found. Mentions of external tools like WoopSocial and Canva are appropriate for the skill's intended use case.- [PROMPT_INJECTION]: Regarding Category 8 (Indirect Prompt Injection), the skill ingests data from external destination URLs to inform design briefs. Ingestion points: Destination page content; Boundary markers: None explicitly defined for untrusted content interpolation; Capability inventory: Text-based design briefing, variant planning, and mockup critique; Sanitization: Not explicitly documented. While an injection surface exists, the agent lacks high-risk capabilities (e.g., file-system writes or unauthorized network requests) to exploit it, resulting in a safe assessment.

pinterest-pin-design