The Agent Skills Directory

[SAFE]: The skill consists entirely of documentation (Markdown files) providing engineering standards. It does not contain any executable scripts, automated commands, or network-active components.
[SAFE]: The guidelines explicitly promote security best practices, such as redacting secrets (e.g., using Redacted<T> wrappers) and minimizing the logging of personal data in references/sensitive-data-and-observability.md.
[SAFE]: The skill emphasizes the 'Parse, don't validate' pattern in references/parsing-and-schemas.md, instructing the agent to treat all external and serialized data as untrusted until it is parsed into domain-specific types, which is a key defense against various injection and data-integrity attacks.
[SAFE]: No obfuscation, prompt injection attempts, persistence mechanisms, or unauthorized privilege escalation commands were detected in the provided content.

coding-standards