The Agent Skills Directory

[SAFE]: The skill implements its functionality using standard Python modules (smtplib, email) without external dependencies or obfuscation.
[DATA_EXFILTRATION]: While the skill can send local files as attachments to external addresses, this is its primary intended function. Credentials are not hardcoded and are managed through a secure key store tool.
[PROMPT_INJECTION]: The skill has a vulnerability surface for indirect prompt injection as it processes untrusted email content and has file-reading capabilities. However, it does not interpret this content as instructions.
Ingestion points: The scripts/send.py script accepts subject, body, and attachment as command-line arguments in SKILL.md.
Boundary markers: No delimiters or explicit instructions are provided to the agent to ignore instructions embedded within the email body.
Capability inventory: The script can read any file path provided to the --attachment argument and transmit data via SMTP.
Sanitization: No input validation or sanitization is performed on the arguments passed to the script.

send-email