The Agent Skills Directory

[SAFE]: The skill consists exclusively of markdown documentation and instructions with no embedded scripts, binaries, or executable code, which significantly limits the attack surface.
[SAFE]: The skill follows security best practices by including explicit instructions to treat all external data (such as scraped text, reviews, and emails) as untrusted evidence rather than instructions, effectively mitigating potential indirect prompt injection attacks.
[SAFE]: The 'Unified Skill Contract' and tool guidelines emphasize credential safety and the principle of least privilege, specifically stating that API keys must be kept out of source control and that persistent writes require explicit user authorization.

popup-cro