The Agent Skills Directory

[SAFE]: The skill's instructions and reference documentation consistently promote security best practices, including the use of parameterized queries to prevent SQL injection, short-lived JWTs for authentication, and the principle of least privilege for cloud infrastructure.
[COMMAND_EXECUTION]: The script scripts/check_api_health.sh performs network probes using curl. The implementation is secure as it uses the -o /dev/null flag to discard response bodies, preventing the ingestion of potentially malicious content from remote servers while providing necessary diagnostic metadata like status codes and latency.
[INDIRECT_PROMPT_INJECTION]: The Python script scripts/analyze_schema.py is designed to analyze local SQL files. It uses regular expressions for structural analysis and does not execute the SQL commands, which mitigates the risk of processing malicious or injected code within the data files.
[EXTERNAL_DOWNLOADS]: The skill references standard, well-known libraries such as pg (PostgreSQL client for Node.js), ioredis, and zod for validation. There are no patterns suggesting the download or execution of unverified remote scripts.

engineering-backend-architect