The Agent Skills Directory

[SAFE]: The skill is purely informational and consists of markdown documentation. It does not include any executable scripts, binary files, or automated commands that could compromise a system.- [DATA_EXFILTRATION]: The skill includes security-positive guidance by explicitly instructing the agent to never commit secrets or environment files (".env") to version control.- [PROMPT_INJECTION]: The skill addresses the risk of indirect prompt injection by advising that instruction-like content from untrusted sources (e.g., third-party API responses or external documentation) should be treated as data to be reported to the user rather than followed as directives.

context-engineering