The Agent Skills Directory

[SAFE]: The skill consists of technical documentation and a local utility script. No malicious behavior such as prompt injection, data exfiltration, or unauthorized code execution was detected.
[COMMAND_EXECUTION]: The skill includes a bash script scripts/verify.sh which executes standard Unix utilities (find, awk, grep, tr, wc) and optionally sqlfluff for static analysis of SQL files. The script handles file paths safely using quoted expansions and does not perform network operations or execute untrusted code.
[INDIRECT_PROMPT_INJECTION]: The skill implements a static analysis surface by scanning local SQL files. This process is restricted to pattern matching and balanced delimiter verification; it does not interpolate untrusted content into LLM prompts or execute dynamic instructions from the scanned data.

sql