The Agent Skills Directory

[SAFE]: No prompt injection patterns or attempts to override agent behavior were detected in the skill instructions or metadata.
[SAFE]: All external URLs provided in the reference lists point to legitimate software directory and review platforms, with no suspicious or malicious domains identified.
[SAFE]: Command-line examples, such as 'curl', are included solely as manual verification steps for the user to perform and are not executed by the agent.
[SAFE]: No obfuscation techniques, hidden content, or data exfiltration mechanisms were found in any of the skill's files.
[SAFE]: The skill processes user-provided product information to generate marketing text; however, this information is not used to construct executable commands or influence secure operations, posing no significant indirect injection risk.

directory-submissions