The Agent Skills Directory

[SAFE]: The skill interacts with the well-known service provider Frontegg. All network operations are directed to the official domain api.frontegg.com and use platform-managed credentials ($FRONTEGG_VENDOR_TOKEN).
[PROMPT_INJECTION]: The skill processes data from external API responses, including user emails, tenant names, and audit logs. This constitutes a potential surface for indirect prompt injection where malicious content in those fields could attempt to influence the agent's subsequent actions.
Ingestion points: SKILL.md (via the frontegg_api helper function and Phase 1/Phase 2 scripts)
Boundary markers: None identified in the prompt templates
Capability inventory: Shell command execution via curl and data processing via jq
Sanitization: The skill uses jq to extract specific fields, which provides structural isolation, but it does not perform content-level sanitization of the strings retrieved from the API.

managing-frontegg