aj/Breehavior-Monitor
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

feat: require warning before mute + sustained toxicity escalation

Browse Source 
Gate mutes behind a prior warning — first offense always gets a warning,
mute only fires if warned_since_reset is True. Warned flag is persisted
to DB (new Warned column on UserState) and survives restarts.

Add post-warning escalation boost to drama_score: each high-scoring
message after a warning adds +0.04 (configurable) so sustained bad
behavior ramps toward the mute threshold instead of plateauing.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
This commit is contained in:
AJ Isaacs
2026-02-25 11:07:57 -05:00
parent f02a4ab49d
commit 71c7b45e9a
4 changed files with 56 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
cogs/sentiment.py
View File

@@ -292,7 +292,8 @@ class SentimentCog(commands.Cog):
# Track the result in DramaTracker
self.bot.drama_tracker.add_entry(user_id, score, categories, reasoning)
drama_score = self.bot.drama_tracker.get_drama_score(user_id)
escalation_boost = sentiment_config.get("escalation_boost", 0.04)
drama_score = self.bot.drama_tracker.get_drama_score(user_id, escalation_boost=escalation_boost)
logger.info(
"User %s (%d) | msg_score=%.2f | drama_score=%.2f | categories=%s | %s",
@@ -358,10 +359,16 @@ class SentimentCog(commands.Cog):
mute_threshold = self.bot.drama_tracker.get_mute_threshold(
user_id, base_mute_threshold
)
user_data = self.bot.drama_tracker.get_user(user_id)
# Mute: rolling average OR single message spike
if drama_score >= mute_threshold or score >= spike_mute:
effective_score = max(drama_score, score)
await self._mute_user(user_ref_msg, effective_score, categories, db_message_id)
if user_data.warned_since_reset:
await self._mute_user(user_ref_msg, effective_score, categories, db_message_id)
else:
# Downgrade to warning — require a warning before muting
logger.info("Downgrading mute to warning for %s (no prior warning)", user_ref_msg.author)
await self._warn_user(user_ref_msg, effective_score, db_message_id)
# Warn: rolling average OR single message spike
elif drama_score >= warning_threshold or score >= spike_warn:
effective_score = max(drama_score, score)
@@ -556,7 +563,8 @@ class SentimentCog(commands.Cog):
self._mark_analyzed(m.id)
self.bot.drama_tracker.add_entry(user_id, score, categories, reasoning)
drama_score = self.bot.drama_tracker.get_drama_score(user_id)
escalation_boost = sentiment_config.get("escalation_boost", 0.04)
drama_score = self.bot.drama_tracker.get_drama_score(user_id, escalation_boost=escalation_boost)
# Save to DB
content_summary = f"[Mention scan] {worst_msg}" if worst_msg else "[Mention scan] See conversation"
@@ -599,9 +607,14 @@ class SentimentCog(commands.Cog):
mute_threshold = self.bot.drama_tracker.get_mute_threshold(
user_id, base_mute_threshold
)
user_data = self.bot.drama_tracker.get_user(user_id)
if drama_score >= mute_threshold or score >= spike_mute:
effective_score = max(drama_score, score)
await self._mute_user(ref_msg, effective_score, categories, db_message_id)
if user_data.warned_since_reset:
await self._mute_user(ref_msg, effective_score, categories, db_message_id)
else:
logger.info("Downgrading mute to warning for %s (no prior warning)", ref_msg.author)
await self._warn_user(ref_msg, effective_score, db_message_id)
elif drama_score >= warning_threshold or score >= spike_warn:
effective_score = max(drama_score, score)
await self._warn_user(ref_msg, effective_score, db_message_id)
@@ -747,6 +760,8 @@ class SentimentCog(commands.Cog):
message_id=db_message_id,
details=f"score={score:.2f}",
))
# Persist warned flag immediately so it survives restarts
self._save_user_state(message.author.id)
async def _handle_topic_drift(
self, message: discord.Message, topic_category: str, topic_reasoning: str,
@@ -897,6 +912,7 @@ class SentimentCog(commands.Cog):
off_topic_count=user_data.off_topic_count,
baseline_coherence=user_data.baseline_coherence,
user_notes=user_data.notes or None,
warned=user_data.warned_since_reset,
))
self._dirty_users.discard(user_id)
@@ -923,6 +939,7 @@ class SentimentCog(commands.Cog):
off_topic_count=user_data.off_topic_count,
baseline_coherence=user_data.baseline_coherence,
user_notes=user_data.notes or None,
warned=user_data.warned_since_reset,
)
logger.info("Flushed %d dirty user states to DB.", len(dirty))

3
config.yaml
View File

@@ -17,8 +17,9 @@ sentiment:
context_messages: 8 # Number of previous messages to include as context
rolling_window_size: 10 # Number of messages to track per user
rolling_window_minutes: 15 # Time window for tracking
batch_window_seconds: 10 # Wait this long for more messages before analyzing (debounce)
batch_window_seconds: 4 # Wait this long for more messages before analyzing (debounce)
escalation_threshold: 0.25 # Triage toxicity score that triggers re-analysis with heavy model
escalation_boost: 0.04 # Per-message drama boost after warning (sustained toxicity ramps toward mute)
game_channels:
gta-online: "GTA Online"

26
utils/database.py
View File

@@ -126,6 +126,12 @@ class Database:
ALTER TABLE UserState ADD UserNotes NVARCHAR(MAX) NULL
""")
# --- Schema migration for warned flag (require warning before mute) ---
cursor.execute("""
IF COL_LENGTH('UserState', 'Warned') IS NULL
ALTER TABLE UserState ADD Warned BIT NOT NULL DEFAULT 0
""")
cursor.execute("""
IF NOT EXISTS (SELECT * FROM sys.tables WHERE name = 'BotSettings')
CREATE TABLE BotSettings (
@@ -284,19 +290,20 @@ class Database:
off_topic_count: int,
baseline_coherence: float = 0.85,
user_notes: str | None = None,
warned: bool = False,
) -> None:
"""Upsert user state (offense count, immunity, off-topic count, coherence baseline, notes)."""
"""Upsert user state (offense count, immunity, off-topic count, coherence baseline, notes, warned)."""
if not self._available:
return
try:
await asyncio.to_thread(
self._save_user_state_sync,
user_id, offense_count, immune, off_topic_count, baseline_coherence, user_notes,
user_id, offense_count, immune, off_topic_count, baseline_coherence, user_notes, warned,
)
except Exception:
logger.exception("Failed to save user state")
def _save_user_state_sync(self, user_id, offense_count, immune, off_topic_count, baseline_coherence, user_notes):
def _save_user_state_sync(self, user_id, offense_count, immune, off_topic_count, baseline_coherence, user_notes, warned):
conn = self._connect()
try:
cursor = conn.cursor()
@@ -306,14 +313,14 @@ class Database:
ON target.UserId = source.UserId
WHEN MATCHED THEN
UPDATE SET OffenseCount = ?, Immune = ?, OffTopicCount = ?,
BaselineCoherence = ?, UserNotes = ?,
BaselineCoherence = ?, UserNotes = ?, Warned = ?,
UpdatedAt = SYSUTCDATETIME()
WHEN NOT MATCHED THEN
INSERT (UserId, OffenseCount, Immune, OffTopicCount, BaselineCoherence, UserNotes)
VALUES (?, ?, ?, ?, ?, ?);""",
INSERT (UserId, OffenseCount, Immune, OffTopicCount, BaselineCoherence, UserNotes, Warned)
VALUES (?, ?, ?, ?, ?, ?, ?);""",
user_id,
offense_count, 1 if immune else 0, off_topic_count, baseline_coherence, user_notes,
user_id, offense_count, 1 if immune else 0, off_topic_count, baseline_coherence, user_notes,
offense_count, 1 if immune else 0, off_topic_count, baseline_coherence, user_notes, 1 if warned else 0,
user_id, offense_count, 1 if immune else 0, off_topic_count, baseline_coherence, user_notes, 1 if warned else 0,
)
cursor.close()
finally:
@@ -356,7 +363,7 @@ class Database:
try:
cursor = conn.cursor()
cursor.execute(
"SELECT UserId, OffenseCount, Immune, OffTopicCount, BaselineCoherence, UserNotes FROM UserState"
"SELECT UserId, OffenseCount, Immune, OffTopicCount, BaselineCoherence, UserNotes, Warned FROM UserState"
)
rows = cursor.fetchall()
cursor.close()
@@ -368,6 +375,7 @@ class Database:
"off_topic_count": row[3],
"baseline_coherence": float(row[4]),
"user_notes": row[5] or "",
"warned": bool(row[6]),
}
for row in rows
]

18
utils/drama_tracker.py
View File

@@ -70,7 +70,7 @@ class DramaTracker:
user.last_analysis_time = now
self._prune_entries(user, now)
def get_drama_score(self, user_id: int) -> float:
def get_drama_score(self, user_id: int, escalation_boost: float = 0.04) -> float:
user = self.get_user(user_id)
now = time.time()
self._prune_entries(user, now)
@@ -86,7 +86,19 @@ class DramaTracker:
weighted_sum += entry.toxicity_score * weight
total_weight += weight
return weighted_sum / total_weight if total_weight > 0 else 0.0
base_score = weighted_sum / total_weight if total_weight > 0 else 0.0
# Escalation: if warned, each high-scoring message AFTER the warning
# adds a boost so sustained bad behavior ramps toward mute threshold
if user.warned_since_reset and user.last_warning_time > 0:
post_warn_high = sum(
1 for e in user.entries
if e.timestamp > user.last_warning_time and e.toxicity_score >= 0.5
)
if post_warn_high > 0:
base_score += escalation_boost * post_warn_high
return min(base_score, 1.0)
def get_mute_threshold(self, user_id: int, base_threshold: float) -> float:
"""Lower the mute threshold if user was already warned."""
@@ -272,6 +284,8 @@ class DramaTracker:
user.baseline_coherence = state["baseline_coherence"]
if "user_notes" in state and state["user_notes"]:
user.notes = state["user_notes"]
if state.get("warned"):
user.warned_since_reset = True
count += 1
return count