The Agent Skills Directory

[SAFE]: The skill does not contain any malicious patterns, obfuscation, or unauthorized access to sensitive data.
[REMOTE_CODE_EXECUTION]: The skill describes how to execute model inference using the runware-run and runware-models tools. These are recognized vendor-provided resources for the Runware platform and are used for the skill's primary intended purpose.
[PROMPT_INJECTION]: The skill processes user-supplied text to generate audio. While it uses specific speaker tags as structural markers, it does not explicitly sanitize the input for embedded instructions. However, the output is limited to synthesized speech, and the risk of significant bypass is minimal.

dialogue-audio