The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill provides patterns for processing external data by interpolating it into email templates, which creates a potential surface for indirect injection if the source data is attacker-controlled.
Ingestion points: User-supplied variables such as userName in the React email component and resetToken in the transactional queue example.
Boundary markers: The code snippets do not demonstrate the use of delimiters or specific instructions to the agent to ignore embedded commands in the interpolated data.
Capability inventory: The skill utilizes network operations via email provider APIs (e.g., resend.emails.send) and database write operations for logging (db.emailLog.create).
Sanitization: No explicit sanitization, HTML escaping, or validation logic is shown for the data being injected into the Html components or transactional payloads.
[EXTERNAL_DOWNLOADS]: The skill references standard, well-known Node.js libraries required for building and processing email content.
Mentions @react-email/components for template construction.
Suggests html-to-text as a utility for generating plain-text fallbacks from HTML content.

email-systems