feat(gateway:moderationAgent): Update rule 8 guidelines

1 files changed, 6 insertions, 2 deletions

diff --git a/packages/gateway/src/listeners/moderationAgent/utilities.ts b/packages/gateway/src/listeners/moderationAgent/utilities.ts
index aa0cbbc..36e90cf 100644
--- a/packages/gateway/src/listeners/moderationAgent/utilities.ts
+++ b/packages/gateway/src/listeners/moderationAgent/utilities.ts
@@ -146,7 +146,7 @@ CRITICAL INSTRUCTIONS:
 RULE-SPECIFIC GUIDELINES:
 - Rule 3 (Spoilers): ONLY flag if media spoilers (movies, TV, anime, manga, etc.) are not properly tagged with ||spoiler|| format. General terms like "NTRd" are NOT spoilers unless they specifically spoil plot details of media.
 - Rule 6 (Underage): ONLY flag if content explicitly states a character is underage OR heavily implies a character is underage. This rule is SPECIFICALLY about sexualizing underage characters/people. Adult threats, adult sexual content, general violence between adults, references to adult family members (mother, father, mom, dad, etc.), or general sexual content between adults is NOT a violation of this rule. Adults can be mothers, fathers, parents - being a parent does NOT imply underage status.
-- Rule 8 (NSFW): ONLY flag if TRULY GRAPHIC sexual content appears in NON-NSFW channels. Check the NSFW Status in the context - if it says "NSFW Channel", then Rule 8 does NOT apply. Non-graphic sexual content, sexual jokes, sexual threats, or general sexual language is allowed in non-NSFW channels.
+- Rule 8 (NSFW): ONLY flag if TRULY GRAPHIC sexual content appears in NON-NSFW channels. Check the NSFW Status in the context - if it says "NSFW Channel", then Rule 8 does NOT apply. Non-graphic sexual content, sexual jokes, sexual threats, sexual violence, or general sexual language is allowed in non-NSFW channels. Rule 8 is about graphic sexual content, NOT threats or violence.
 - Rule 13 (Test String): Only flag if the exact string "test_string_1337" appears
 
 WHAT RULE 6 DOES NOT COVER:
@@ -165,6 +165,7 @@ WHAT RULE 6 DOES NOT COVER:
 WHAT RULE 8 DOES NOT COVER (NON-GRAPHIC CONTENT ALLOWED IN SFW CHANNELS):
 - Sexual jokes or humor
 - Sexual threats or language
+- Sexual violence or threats of sexual violence
 - General sexual references
 - Crude sexual language
 - Sexual insults or taunts
@@ -173,6 +174,7 @@ WHAT RULE 8 DOES NOT COVER (NON-GRAPHIC CONTENT ALLOWED IN SFW CHANNELS):
 - General sexual banter
 - Sexual roleplay references
 - Sexual memes or jokes
+- Threats of rape or sexual violence (these are threats, not graphic sexual content)
 
 EXAMPLE: "I will fucking rape you nigga" is NOT a Rule 6 violation because:
 - It's a threat between adults
@@ -196,6 +198,7 @@ FALSE FLAG EXAMPLES - THESE ARE NOT VIOLATIONS:
 - "your own father" - NOT Rule 6 violation (adults have fathers, no age indication)
 - "UMA PUSSY HERE I COME" - NOT Rule 8 violation (not graphic sexual content)
 - "I want to fuck Special Weeks uma mom" - NOT Rule 6 violation (adults can be mothers, no age indication)
+- "Where do I find the nearest cockroach woman to rape?" - NOT Rule 8 violation (threat/violence, not graphic sexual content)
 
 RULE 8 NSFW CHECK:
 - ALWAYS check the "NSFW Status" field in the context
@@ -206,7 +209,8 @@ RULE 8 NSFW CHECK:
 
 WHAT IS NOT GRAPHIC SEXUAL CONTENT (ALLOWED IN SFW CHANNELS):
 - Sexual insults ("go fuck yourself", "fuck you")
-- Sexual threats ("I'll fuck your wife")
+- Sexual threats ("I'll fuck your wife", "I'll rape you")
+- Threats of sexual violence (rape threats, sexual assault threats)
 - General sexual language ("pussy", "fuck", "sex")
 - Sexual jokes or humor
 - Sexual roleplay references