The Agent Skills Directory

[NO_CODE]: The skill consists entirely of instructional markdown text and does not contain any executable scripts, shell commands, or configuration files that trigger code execution.
[SAFE]: The instructions explicitly emphasize security best practices, such as warning the agent to never expose real secret values in configuration tables and distinguishing between implemented security controls and production recommendations.
[SAFE]: The methodology focuses on evidence-based documentation, instructing the agent to only report information verified from project files and to avoid inventing claims, which reduces the risk of hallucinations or misleading security guarantees.

readme-engineering