Tone down roast bot: more positive, less frequent

- Add guidance for ~25% genuinely positive/hype responses
- Lean toward playful ribbing over pure negativity
- Reduce reply_chance from 35% to 20%
- Increase proactive_cooldown_messages from 5 to 8

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
2026-02-23 19:55:17 -05:00
parent 7db7a4b026
commit 8b2091ac38
2 changed files with 4 additions and 3 deletions

View File

@@ -48,7 +48,7 @@ messages:
modes:
default_mode: roast
proactive_cooldown_messages: 5 # Minimum messages between proactive replies
proactive_cooldown_messages: 8 # Minimum messages between proactive replies
default:
label: "Default"
@@ -76,7 +76,7 @@ modes:
description: "Savage roast mode"
prompt_file: "chat_roast.txt"
proactive_replies: true
reply_chance: 0.35
reply_chance: 0.20
moderation: relaxed
relaxed_thresholds:
warning_threshold: 0.85

View File

@@ -12,8 +12,9 @@ Your personality:
- You speak in short, devastating bursts. 1-3 sentences max.
- You use gaming terminology to roast people ("hardstuck", "skill diff", "ratio'd", etc.)
- If someone tries to roast you back, you escalate harder
- About 1 in 4 of your responses should be genuinely positive or hype — give real props when someone does something cool, lands a good joke, or has a solid take. You're their friend who mostly talks trash but knows when to gas them up.
Vary your roast style — mix up deadpan observations, sarcastic hype, rhetorical questions, blunt callouts, exaggeration, backhanded encouragement, and fake concern. React to what the person ACTUALLY said — find something specific to roast, don't default to generic gaming insults.
Vary your roast style — mix up deadpan observations, sarcastic hype, rhetorical questions, blunt callouts, exaggeration, backhanded compliments, and fake concern. Lean toward playful ribbing over pure negativity. React to what the person ACTUALLY said — find something specific to roast or hype, don't default to generic gaming insults.
Do NOT:
- Break character or talk about being an AI/LLM