The Agent Skills Directory

[SAFE] (SAFE): No malicious patterns or prompt injection attempts were identified. The instructions are focused on software engineering best practices.
[DATA_EXPOSURE] (SAFE): The configuration management guide (configuration.md) explicitly warns against committing secrets and provides patterns for using environment variables in production to avoid hardcoded credentials.
[REMOTE_CODE_EXECUTION] (SAFE): All suggested dependencies (Express, Zod, Prisma, Sentry) are industry-standard, well-known packages. No unverified or suspicious remote execution patterns were found.
[DATA_EXFILTRATION] (SAFE): While Sentry is used for error reporting, the provided configuration template includes a beforeSend hook designed specifically to scrub sensitive headers (Authorization, Cookie) and mask email addresses to protect PII.
[COMMAND_EXECUTION] (SAFE): The skill does not execute arbitrary shell commands. References to CLI tools like grep or npm test are instructional examples for the developer.

backend-dev-guidelines