The Agent Skills Directory

[NO_CODE] (SAFE): The skill contains only markdown instructions and metadata. No scripts, binaries, or automation code are included.
[DATA_EXFILTRATION] (SAFE): No commands or patterns were found that access credentials, sensitive environment variables, or perform external network requests.
[INDIRECT_PROMPT_INJECTION] (LOW): The skill is designed to process untrusted data, specifically SQL queries and database logs. While this creates an ingestion surface for indirect prompt injection, the lack of executable capabilities in the skill's own code limits the risk to the agent's internal reasoning.

database-optimizer