The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes npx to fetch and execute the resend-mcp package from the npm registry. This is the standard method for running the Resend MCP server.
[DATA_EXFILTRATION]: The send_email and batch_send_emails tools include an attachments feature that can read from a local file path. This capability allows the agent to send local file content to external recipients through the Resend API. This is an intended functionality but presents a surface for data exfiltration if the agent is misdirected.
[PROMPT_INJECTION]: The skill processes external data for email bodies and subjects, which introduces a risk of indirect prompt injection.
Ingestion points: The from, to, subject, and body fields in the send_email and batch_send_emails operations (SKILL.md).
Boundary markers: No specific delimiters or safety instructions are defined to separate untrusted content from agent instructions.
Capability inventory: File system access via the path parameter for attachments and network communication via the Resend API.
Sanitization: No explicit sanitization or validation of the email content is mentioned in the skill definition.

resend