The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data via bulk email lists (CSVs), which creates an ingestion surface for potential indirect prompt injection attacks where malicious instructions could be embedded in user-uploaded content.
Ingestion points: User-provided CSV files processed by the Bulk Validation API endpoints described in SKILL.md.
Boundary markers: The 'Security' section explicitly warns the agent to treat downloaded validation results as untrusted content.
Capability inventory: The skill facilitates network operations to the Mailgun API and document fetching from trusted documentation sites.
Sanitization: Instructions mandate that the agent validate and sanitize email addresses and metadata before processing or storage to mitigate risks from untrusted user content.
[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch authoritative technical documentation and API schemas from official vendor domains.
Evidence: Network requests and documentation links targeting documentation.mailgun.com and developers.sinch.com.
[SAFE]: The skill includes robust security guidance for handling API keys, recommending environment variables over hardcoding, and emphasizes PII protection for email lists.

sinch-mailgun-validate