The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were detected in the provided instructions or code snippets. The skill serves its stated purpose as a developer guide for email infrastructure.
[CREDENTIALS_UNSAFE]: The skill demonstrates best practices for secret management by referencing environment variables (e.g., process.env.RESEND_API_KEY, process.env.REDIS_URL) instead of hardcoding sensitive credentials.
[EXTERNAL_DOWNLOADS]: All referenced Node.js packages (resend, @react-email/components, bullmq, ioredis) are standard, well-known libraries for email and task queue management.
[INDIRECT_PROMPT_INJECTION]: The skill provides a surface for data ingestion via email template variables. While this is an inherent property of templating systems, the skill follows standard implementation patterns.
Ingestion points: WelcomeEmailProps in SKILL.md accepts user-supplied strings such as userName.
Boundary markers: Absent (standard for React-based HTML templating).
Capability inventory: The skill provides logic for network transmission of generated content via the Resend SDK.
Sanitization: The implementation assumes data is sanitized at the application layer before being passed to the templating engine.

email-systems