aiml-moderation-content
Content moderation benchmark with 3 variants: guard toxic text (CSV), guard user input prompts (TXT), guard model output responses (JSONL). Use when: testing ISC on content moderation, generating toxic text samples, attack prompts, or unsafe model responses. Keywords: moderation, toxic, hate speech, jailbreak, harmful compliance.
Installation and usage
Content moderation benchmark with 3 variants: guard toxic text (CSV), guard user input prompts (TXT), guard model output responses (JSONL). Use when: testing ISC on content moderation, generating toxic text samples, attack prompts, or unsafe model responses. Keywords: moderation, toxic, hate speech, jailbreak, harmful compliance.
После установки вы можете использовать этот skill, выполнив следующую команду в терминале:
skills use aiml-moderation-content