Compare commits

...

3 Commits

Author SHA1 Message Date
Alexander Whitestone
d9f53d900a fix: deprecate dying_detection and consolidate crisis detection (#40)
All checks were successful
Sanity Checks / sanity-test (pull_request) Successful in 8s
Smoke Test / smoke (pull_request) Successful in 10s
2026-04-14 11:34:49 -04:00
9b94978d1c feat: Wire compassion router into gateway flow (#34) (#43)
All checks were successful
Smoke Test / smoke (push) Successful in 5s
Squash merge: wire compassion router into gateway flow
2026-04-13 19:59:15 +00:00
e71bca1744 fix: de-duplicate crisis_detector.py and crisis/detect.py (closes #39) (#44)
Some checks failed
Smoke Test / smoke (push) Has been cancelled
Squash merge: de-duplicate crisis detector (closes #39)

Co-authored-by: Alexander Whitestone <alexander@alexanderwhitestone.com>
Co-committed-by: Alexander Whitestone <alexander@alexanderwhitestone.com>
2026-04-13 19:59:12 +00:00
6 changed files with 439 additions and 661 deletions

View File

@@ -1,25 +1,24 @@
"""
Crisis Detection Module for the-door.
Crisis Detection Module for the-door (CANONICAL).
Single source of truth for crisis keyword detection.
Parses incoming text for despair/suicide indicators and classifies into
tiers: LOW, MEDIUM, HIGH, CRITICAL.
Both `crisis/detect.py` and `crisis_detector.py` (legacy shim) use this module.
Do NOT duplicate keyword lists elsewhere.
Consolidated from:
- crisis_detector.py (root, legacy — MODERATE tier, class API)
- crisis/detect.py (module, newer — MEDIUM tier, functional API)
"""
import re
from dataclasses import dataclass, field
from typing import List
from typing import List, Optional
@dataclass
class CrisisDetectionResult:
level: str
indicators: List[str] = field(default_factory=list)
recommended_action: str = ""
score: float = 0.0
matches: List[dict] = field(default_factory=list)
# ── Indicator sets ──────────────────────────────────────────────
# ── Canonical indicator sets ──────────────────────────────────────
CRITICAL_INDICATORS = [
r"\bbetter off without me\b",
@@ -29,49 +28,58 @@ CRITICAL_INDICATORS = [
r"\bnot\s+worth\s+living\b",
r"\bbetter\s+off\s+dead\b",
r"\bend\s+it\s+all\b",
r"\bcan'?t\s+(go|live)\s+on\b",
r"\bno\s+reason\s+to\s+live\b",
r"\bdon'?t\s+want\s+to\s+live\b",
r"\bwant\s+to\s+die\b",
r"\bgoing\s+to\s+kill\s+myself\b",
r"\bplan\s+to\s+(end|kill|die)\b",
r"\btired\s+of\s+(living|life|existence)\b",
r"\bsaying\s+goodbye\s+(forever|permanently|one last time)\b",
r"\bgoing\s+to\s+(?:kill\s+myself|die)\b",
r"\bplan\s+to\s+(?:end|kill|die)\b",
r"\btired\s+of\s+(?:living|life|existence)\b",
r"\bsaying\s+goodbye\s+(?:forever|permanently|one\s+last\s+time)\b",
r"\bwrote\s+a\s+suicide\s*(?:note|letter)\b",
r"\bgiving\s+away\s+(?:my|all\s+my)\s+(?:stuff|things|possessions?)\s+(?:to|because|—)\b",
r"\btied\s+(up|down)\s+my\s+(loose\s+)?ends",
r"\btied\s+(?:up|down)\s+my\s+(?:loose\s+)?ends",
]
HIGH_INDICATORS = [
r"\bdespair\b",
r"\bhopeless(?:ly)?\s+(about\s+(?:my|this|everything|life)|inside|right\s+now)\b",
r"\bno(?!t)\s+(one|body|point|hope|way\s+out)\b",
r"\bhopeless(?:ly)?\s+(?:about\s+(?:my|this|everything|life)|inside|right\s+now)\b",
r"\bno(?!t)\s+(?:one|body|point|hope|way\s+out)\b",
r"\bno\s+future\s+(?:for\s+me|ahead|left)\b",
r"\beverything\s+is\s+(pointless|broken|ruined)\b",
r"\beverything\s+is\s+(?:pointless|broken|ruined)\b",
r"\bcan'?t\s+take\s+this\s+anymore\b",
r"\bdon'?t\s+care\s+if\s+I\s+die\b",
r"\bwish\s+I\s+(was|were)\s+(dead|gone|never\s+born)\b",
r"\bwish\s+I\s+(?:was|were)\s+(?:dead|gone|never\s+born)\b",
r"\bdon'?t\s+matter\s+if\s+I\s+exist\b",
r"\bno\s+one\s+would\s+care\b",
r"\bno\s+one\s+would\s+miss\b",
r"\bworld\s+would?\s+be\s+better\s+without\b",
r"\bin\s+so\s+much\s+(pain|agony|suffering|torment)\b",
r"\bcan'?t\s+see\s+any\s+(point|reason)\b",
r"\bno\s+one\s+would\s+(?:care|miss)\b",
r"\bworld\s+would?\s+be\s+better\s+without\s+me\b",
r"\bin\s+so\s+much\s+(?:pain|agony|suffering|torment|anguish)\b",
r"\bcan'?t\s+see\s+any\s+(?:point|reason|hope|way)\b",
r"\bescape\s+from\s*this",
r"\bjust\s+want\s+it\s+to\s+stop\b",
r"\btrapped\s+(?:in\s+(?:my|this|a\s+dark)|and\s+can'?t\s+escape)\b",
r"\bnothing\s+left\s+(?:to\s+(?:live\s+for|hope\s+for|give)|inside)\b",
r"\bdisappeared\s+forever\b",
# Contextual despair phrases (from crisis_detector.py legacy)
r"\bfeel(?:s|ing)?\s+(?:so\s+)?hopeless\b",
r"\beverything\s+is\s+hopeless\b",
r"\bcan'?t\s+(?:go\s+on|keep\s+going)\b",
r"\bgive(?:n)?\s*up\s+(?:on\s+)?(?:life|living|everything)\b",
r"\bgive(?:n)?\s*up\s+on\s+myself\b",
r"\bno\s*point\s+(?:in\s+)?living\b",
r"\bno\s*hope\s+(?:left|remaining)\b",
r"\bno\s*way\s*out\b",
r"\bfeel(?:s|ing)?\s+trapped\b",
r"\btrapped\s+in\s+this\s+(?:situation|life|pain|darkness|hell)\b",
r"\btrapped\s+and\s+can'?t\s+escape\b",
r"\bdesperate\s+(?:for\s+)?help\b",
r"\bfeel(?:s|ing)?\s+desperate\b",
]
MEDIUM_INDICATORS = [
r"\bno\s+hope\b",
r"\bcan'?t\s+go\s+on\b",
r"\bcan'?t\s+keep\s+going\b",
r"\bforgotten\b",
r"\balone\s+in\s+this\b",
r"\balways\s+alone\b",
r"\bnobody\s+understands\b",
r"\bnobody\s+cares\b",
r"\bnobody\s+(?:understands|cares)\b",
r"\bwish\s+I\s+could\b",
r"\bexhaust(?:ed|ion|ing)\b",
r"\bnumb\b",
@@ -80,8 +88,7 @@ MEDIUM_INDICATORS = [
r"\buseless\b",
r"\bbroken\b",
r"\bdark(ness)?\b",
r"\bdepressed\b",
r"\bdepression\b",
r"\bdepress(?:ed|ion)\b",
r"\bcrying\b",
r"\btears\b",
r"\bsad(ness)?\b",
@@ -89,129 +96,119 @@ MEDIUM_INDICATORS = [
r"\boverwhelm(?:ed|ing)\b",
r"\bfailing\b",
r"\bcannot\s+cope\b",
r"\blosing\s*(my)?\s*control\b",
r"\blosing\s*(?:my)?\s*control\b",
r"\bdown\s*for\s*the\s*count\b",
r"\bsinking\b",
r"\bdrowning\b",
r"\bhopeless\b",
# Removed 'desperate' - too broad for MEDIUM, triggers on competition
r"\blost\s+all\s+hope\b",
r"\bno\s+tomorrow\b",
# Contextual versions (from crisis_detector.py legacy)
r"\bfeel(?:s|ing)?\s+(?:so\s+)?worthless\b",
r"\bfeel(?:s|ing)?\s+(?:so\s+)?hopeless\b",
r"\bfeel(?:s|ing)?\s+trapped\b",
r"\bfeel(?:s|ing)?\s+desperate\b",
r"\bno\s+future\s+(?:for\s+me|ahead|left)\b",
r"\bnothing\s+left\s+(?:to\s+(?:live|hope)\s+for|inside)\b",
r"\bgive(?:n)?\s*up\s+on\s+myself\b",
]
LOW_INDICATORS = [
r"\bunhappy\b",
r"\bdown\b",
r"\btough\s*time\b",
r"\brough\s*day\b",
r"\brough\s*week\b",
r"\brough\s*patch\b",
r"\btough\s*(?:time|day|week)\b",
r"\brough\s+(?:day|week|patch)\b",
r"\bstressed\b",
r"\burnout\b",
r"\bburnout\b",
r"\bdifficult\b",
r"\bfrustrated\b",
r"\bannoyed\b",
r"\btired\b",
r"\bsad\b",
r"\bupset\b",
r"\bnot\s*great\b",
r"\bnot\s*good\b",
r"\bnot\s*okay\b",
r"\blonely\b",
r"\banxious?\b",
r"\bnot\s*(?:good|great|okay)\b",
r"\bthings\s*are\s*hard\b",
r"\bstruggling\b",
r"\bnot\s+feeling\s+myself\b",
r"\bnot\s+feeling\s+(?:like\s+)?myself\b",
r"\bhard\s*time\b",
]
def detect_crisis(text: str) -> CrisisDetectionResult:
# ── Scoring ───────────────────────────────────────────────────────
SCORES = {"CRITICAL": 1.0, "HIGH": 0.75, "MEDIUM": 0.5, "LOW": 0.25, "NONE": 0.0}
ACTIONS = {
"CRITICAL": (
"INTERRUPT CONVERSATION. Ask 'Are you safe right now?' "
"Provide 988 Suicide & Crisis Lifeline. "
"Provide Crisis Text Line (HOME to 741741). "
"Stay present. Do not disconnect. Redirect to professional help."
),
"HIGH": (
"Show crisis panel. Ask about safety. Surface 988 number prominently. "
"Continue conversation with crisis awareness."
),
"MEDIUM": (
"Increase warmth and presence. Subtly surface help resources. "
"Keep conversation anchored in the present."
),
"LOW": (
"Normal conversation with warm undertone. "
"No crisis UI elements needed. Remain vigilant."
),
}
# ── Result types ──────────────────────────────────────────────────
@dataclass
class CrisisDetectionResult:
"""Result used by crisis/detect.py, gateway, dying_detection."""
level: str
indicators: List[str] = field(default_factory=list)
recommended_action: str = ""
score: float = 0.0
matches: List[dict] = field(default_factory=list)
@dataclass
class CrisisResult:
"""Legacy result used by crisis_detector.py and crisis_responder.py.
Backward-compatible shim: wraps CrisisDetectionResult with the old field names.
"""
Detect crisis level in a message.
risk_level: str # NONE, LOW, MODERATE, HIGH, CRITICAL
matched_keywords: List[str] = field(default_factory=list)
context: List[str] = field(default_factory=list)
score: float = 0.0
timestamp: Optional[str] = None
Returns:
CrisisDetectionResult with level, found indicators, recommended action, score
"""
if not text or not text.strip():
return CrisisDetectionResult(level="NONE", score=0.0)
def __bool__(self):
return self.risk_level != "NONE"
text_lower = text.lower()
matches = _find_indicators(text_lower)
if not matches:
return CrisisDetectionResult(level="NONE", score=0.0)
# Priority: highest tier wins
# CRITICAL and HIGH: single match is enough
# MEDIUM: requires 2+ indicators (threshold) to reduce false positives
# LOW: single match is enough
scores = {"CRITICAL": 1.0, "HIGH": 0.75, "MEDIUM": 0.5, "LOW": 0.25}
actions = {
"CRITICAL": (
"INTERRUPT CONVERSATION. Ask 'Are you safe right now?' "
"Provide 988 Suicide & Crisis Lifeline. "
"Provide Crisis Text Line (HOME to 741741). "
"Stay present. Do not disconnect. Redirect to professional help."
),
"HIGH": (
"Show crisis panel. Ask about safety. Surface 988 number prominently. "
"Continue conversation with crisis awareness."
),
"MEDIUM": (
"Increase warmth and presence. Subtly surface help resources. "
"Keep conversation anchored in the present."
),
"LOW": (
"Normal conversation with warm undertone. "
"No crisis UI elements needed. Remain vigilant."
),
}
for tier in ("CRITICAL", "HIGH"):
if matches[tier]:
tier_matches = matches[tier]
patterns = [m["pattern"] for m in tier_matches]
return CrisisDetectionResult(
level=tier,
indicators=patterns,
recommended_action=actions[tier],
score=scores[tier],
matches=tier_matches,
)
# MEDIUM tier: require at least 2 indicators before escalating
if len(matches["MEDIUM"]) >= 2:
tier_matches = matches["MEDIUM"]
patterns = [m["pattern"] for m in tier_matches]
return CrisisDetectionResult(
level="MEDIUM",
indicators=patterns,
recommended_action=actions["MEDIUM"],
score=scores["MEDIUM"],
matches=tier_matches,
@classmethod
def from_detection_result(cls, dr: CrisisDetectionResult, text: str = "") -> "CrisisResult":
"""Convert a CrisisDetectionResult to legacy CrisisResult format."""
# Map MEDIUM -> MODERATE for legacy consumers
level = "MODERATE" if dr.level == "MEDIUM" else dr.level
# Extract context snippets from matches
contexts = []
if text:
for m in dr.matches:
ctx = extract_context(text, m["start"], m["end"])
contexts.append(ctx)
return cls(
risk_level=level,
matched_keywords=dr.indicators,
context=contexts,
score=dr.score,
)
if matches["LOW"]:
tier_matches = matches["LOW"]
patterns = [m["pattern"] for m in tier_matches]
return CrisisDetectionResult(
level="LOW",
indicators=patterns,
recommended_action=actions["LOW"],
score=scores["LOW"],
matches=tier_matches,
)
# Single MEDIUM match falls through to LOW sensitivity
if matches["MEDIUM"]:
tier_matches = matches["MEDIUM"]
patterns = [m["pattern"] for m in tier_matches]
return CrisisDetectionResult(
level="LOW",
indicators=patterns,
recommended_action=actions["LOW"],
score=scores["LOW"],
matches=tier_matches,
)
return CrisisDetectionResult(level="NONE", score=0.0)
# ── Core detection ────────────────────────────────────────────────
def _find_indicators(text: str) -> dict:
"""Return dict with indicators found per tier, including match positions."""
@@ -240,6 +237,140 @@ def _find_indicators(text: str) -> dict:
return results
def detect_crisis(text: str) -> CrisisDetectionResult:
"""
Detect crisis level in a message.
Detection hierarchy:
CRITICAL — immediate risk of self-harm or suicide (single match)
HIGH — strong despair signals, ideation present (single match)
MEDIUM — distress signals, requires 2+ indicators to escalate
LOW — emotional difficulty, warrant gentle support (single match)
NONE — no crisis indicators detected
Design principles:
- Never computes the value of a human life
- Never suggests someone should die or that death is a solution
- Always errs on the side of higher risk when uncertain
"""
if not text or not text.strip():
return CrisisDetectionResult(level="NONE", score=0.0)
text_lower = text.lower()
matches = _find_indicators(text_lower)
if not matches:
return CrisisDetectionResult(level="NONE", score=0.0)
# CRITICAL and HIGH: single match is enough
for tier in ("CRITICAL", "HIGH"):
if matches[tier]:
tier_matches = matches[tier]
patterns = [m["pattern"] for m in tier_matches]
return CrisisDetectionResult(
level=tier,
indicators=patterns,
recommended_action=ACTIONS[tier],
score=SCORES[tier],
matches=tier_matches,
)
# MEDIUM tier: require at least 2 indicators before escalating
if len(matches["MEDIUM"]) >= 2:
tier_matches = matches["MEDIUM"]
patterns = [m["pattern"] for m in tier_matches]
return CrisisDetectionResult(
level="MEDIUM",
indicators=patterns,
recommended_action=ACTIONS["MEDIUM"],
score=SCORES["MEDIUM"],
matches=tier_matches,
)
if matches["LOW"]:
tier_matches = matches["LOW"]
patterns = [m["pattern"] for m in tier_matches]
return CrisisDetectionResult(
level="LOW",
indicators=patterns,
recommended_action=ACTIONS["LOW"],
score=SCORES["LOW"],
matches=tier_matches,
)
# Single MEDIUM match falls through to LOW sensitivity
if matches["MEDIUM"]:
tier_matches = matches["MEDIUM"]
patterns = [m["pattern"] for m in tier_matches]
return CrisisDetectionResult(
level="LOW",
indicators=patterns,
recommended_action=ACTIONS["LOW"],
score=SCORES["LOW"],
matches=tier_matches,
)
return CrisisDetectionResult(level="NONE", score=0.0)
# ── CrisisDetector class (backward compat) ───────────────────────
class CrisisDetector:
"""
Legacy class API for crisis detection. Wraps the canonical detect_crisis().
Used by crisis_responder.py and tests/test_false_positive_fixes.py.
Maps MEDIUM -> MODERATE for legacy consumers.
"""
def scan(self, text: str) -> CrisisResult:
dr = detect_crisis(text)
return CrisisResult.from_detection_result(dr, text=text)
def scan_multiple(self, texts: List[str]) -> List[CrisisResult]:
return [self.scan(t) for t in texts]
def get_highest_risk(self, texts: List[str]) -> CrisisResult:
results = self.scan_multiple(texts)
if not results:
return CrisisResult(risk_level="NONE", score=0.0)
return max(results, key=lambda r: r.score)
@staticmethod
def format_result(result: CrisisResult) -> str:
level_emoji = {
"CRITICAL": "\U0001f6a8",
"HIGH": "\u26a0\ufe0f",
"MODERATE": "\U0001f536",
"LOW": "\U0001f535",
"NONE": "\u2705",
}
emoji = level_emoji.get(result.risk_level, "\u2753")
lines = [
f"{emoji} Risk Level: {result.risk_level} (score: {result.score:.2f})",
f"Matched keywords: {len(result.matched_keywords)}",
]
if result.matched_keywords:
lines.append(f" Keywords: {', '.join(result.matched_keywords)}")
if result.context:
lines.append("Context:")
for ctx in result.context:
lines.append(f" {ctx}")
return "\n".join(lines)
# ── Module-level convenience (backward compat) ────────────────────
_default_detector = CrisisDetector()
def detect_crisis_legacy(text: str) -> CrisisResult:
"""Convenience function returning legacy CrisisResult format."""
return _default_detector.scan(text)
# ── Utility functions ─────────────────────────────────────────────
def scan(text: str) -> CrisisDetectionResult:
"""Alias for detect_crisis — shorter name used in tests."""
return detect_crisis(text)
@@ -258,8 +389,8 @@ def extract_context(text: str, start: int, end: int, window: int = 60) -> str:
def get_urgency_emoji(level: str) -> str:
mapping = {"CRITICAL": "🚨", "HIGH": "⚠️", "MEDIUM": "🔶", "LOW": "🔵", "NONE": ""}
return mapping.get(level, "")
mapping = {"CRITICAL": "\U0001f6a8", "HIGH": "\u26a0\ufe0f", "MEDIUM": "\U0001f536", "LOW": "\U0001f535", "NONE": "\u2705"}
return mapping.get(level, "\u2753")
def format_result(result: CrisisDetectionResult) -> str:

View File

@@ -1,268 +1,34 @@
"""
Crisis Detection System for the-door.
Crisis Detection System for the-door (LEGACY SHIM).
Scans text for suicide, self-harm, and hopelessness signals.
Returns risk level, matched keywords, and surrounding context.
This module is a backward-compatible re-export layer.
The canonical detection logic lives in crisis/detect.py.
This is the most important code in the foundation.
Every line here exists because someone's life depends on it.
Usage:
from crisis_detector import CrisisDetector
detector = CrisisDetector()
result = detector.scan("I can't go on anymore")
print(result.risk_level) # "HIGH"
print(result.matched_keywords) # ["can't go on"]
print(result.context) # surrounding text snippets
Both crisis_responder.py and the legacy test suite import from here.
Do NOT add detection logic to this file — it all comes from crisis.detect.
"""
import re
from dataclasses import dataclass, field
from typing import List, Optional, Tuple
from crisis.detect import (
CrisisDetector,
CrisisResult,
CrisisDetectionResult,
detect_crisis_legacy as detect_crisis,
CRITICAL_INDICATORS,
HIGH_INDICATORS,
MEDIUM_INDICATORS,
LOW_INDICATORS,
SCORES as RISK_SCORES,
)
@dataclass
class CrisisResult:
"""Result of a crisis text scan."""
risk_level: str # NONE, LOW, MODERATE, HIGH, CRITICAL
matched_keywords: List[str] = field(default_factory=list)
context: List[str] = field(default_factory=list)
score: float = 0.0
timestamp: Optional[str] = None
def __bool__(self):
return self.risk_level != "NONE"
# ── Keyword definitions ──────────────────────────────────────────
CRITICAL_KEYWORDS = {
"suicide": r"\bsuicid(?:al|e|ed)\b",
"kill myself": r"\bkill\s*(?:my)?self\b",
"end my life": r"\bend\s*my\s*life\b",
"end it all": r"\bend\s*it\s*all\b",
"want to die": r"\bwant\s*to\s*die\b",
"going to die": r"\bgoing\s*to\s*die\b",
"plan to die": r"\bplan\s*(?:to|for)\s*(?:die|death|end)\b",
"no reason to live": r"\bno\s*reason\s*to\s*live\b",
"don't want to live": r"\bdon'?t\s*want\s*to\s*live\b",
"not worth living": r"\bnot\s*worth\s*living\b",
"better off dead": r"\bbetter\s*off\s*dead\b",
"better off without me": r"\bbetter\s*off\s*without\s*me\b",
"goodbye forever": r"\bgoodbye\s*forever\b",
"saying goodbye forever": r"\bsaying\s*goodbye\s*(?:forever|permanently|one\s*last\s*time)\b",
"tired of living": r"\btired\s*of\s*(?:living|life|existence)\b",
"wrote a suicide note": r"\bwrote\s*(?:a|my)\s*(?:suicide\s*note|suicide\s*letter)\b",
"giving away my stuff": r"\bgiving\s*away\s*(?:my|all)\s*(?:stuff|things|possessions?)\s*(?:to|because|—)\b",
}
HIGH_KEYWORDS = {
# Phrases that strongly indicate despair in context
"feel hopeless": r"\bfeel(?:s|ing)?\s+(?:so\s+)?hopeless\b",
"everything is hopeless": r"\beverything\s+is\s+hopeless\b",
"can't go on": r"\bcan'?t\s*go\s*on\b",
"can't keep going": r"\bcan'?t\s*keep\s*going\b",
"can't take this": r"\bcan'?t\s*take\s*this\b",
"give up on life": r"\bgive(?:n)?\s*up\s+(?:on\s+)?(?:life|living|everything)\b",
"give up on myself": r"\bgive(?:n)?\s*up\s+on\s+myself\b",
"no point in living": r"\bno\s*point\s+(?:in\s+)?living\b",
"no hope left": r"\bno\s*hope\s+(?:left|remaining)\b",
"no way out": r"\bno\s*way\s*out\b",
"trapped in this": r"\btrapped\s+(?:in\s+)?(?:this|my|life|situation)\b",
"feel trapped": r"\bfeel(?:s|ing)?\s+trapped\b",
"desperate for help": r"\bdesperate\s+(?:for\s+)?help\b",
"feel desperate": r"\bfeel(?:s|ing)?\s+desperate\b",
"wish I was dead": r"\bwish\s*I\s*(?:was|were)\s*(?:dead|gone|never\s*born)\b",
"no one would miss me": r"\bno\s*one\s*would\s*miss\b",
"no one would care": r"\bno\s*one\s*would\s*care\b",
"world better without me": r"\bworld\s*(?:would|will)\s*be\s*better\s*without\b",
"so much pain": r"\bin\s*so\s*much\s*pain\b",
"can't see any way": r"\bcan'?t\s*see\s*(?:any\s*)?way\b",
"just want it to stop": r"\bjust\s*want\s*it\s*to\s*stop\b",
"don't care if I die": r"\bdon'?t\s*care\s*if\s*I\s*die\b",
}
MODERATE_KEYWORDS = {
"alone": r"\balone\b",
"lost": r"\blost\b",
"broken": r"\bbroken\b",
"afraid": r"\bafraid\b",
"pain": r"\b(?:in\s*)?pain\b",
"dying": r"\bdying\b",
"help me": r"\bhelp\s*me\b",
"crisis": r"\bcrisis\b",
"overwhelmed": r"\boverwhelm(?:ed|ing)\b",
"exhausted": r"\bexhausted\b",
"numb": r"\bnumb\b",
"empty": r"\bempty\b",
"depressed": r"\bdepressed\b",
"depression": r"\bdepression\b",
"despair": r"\bdespair\b",
"miserable": r"\bmiserable\b",
"drowning": r"\bdrowning\b",
"sinking": r"\bsinking\b",
"nobody cares": r"\bnobody\s*cares\b",
"nobody understands": r"\bnobody\s*understands\b",
# Contextual versions of common words
"feel worthless": r"\bfeel(?:s|ing)?\s+(?:so\s+)?worthless\b",
"feel hopeless": r"\bfeel(?:s|ing)?\s+(?:so\s+)?hopeless\b",
"feel trapped": r"\bfeel(?:s|ing)?\s+trapped\b",
"feel desperate": r"\bfeel(?:s|ing)?\s+desperate\b",
"no future for me": r"\bno\s+future\s+(?:for\s+me|ahead|left)\b",
"nothing left to live for": r"\bnothing\s+left\s+(?:to\s+(?:live|hope)\s+for|inside)\b",
"give up on myself": r"\bgive(?:n)?\s*up\s+on\s+myself\b",
}
LOW_KEYWORDS = {
"unhappy": r"\bunhappy\b",
"struggling": r"\bstruggling\b",
"stressed": r"\bstressed\b",
"frustrated": r"\bfrustrated\b",
"tired": r"\btired\b",
"sad": r"\bsad\b",
"upset": r"\bupset\b",
"down": r"\bdown\b",
"tough time": r"\btough\s*time\b",
"rough day": r"\brough\s*day\b",
"rough week": r"\brough\s*week\b",
"rough patch": r"\brough\s*patch\b",
"hard time": r"\bhard\s*time\b",
"difficult": r"\bdifficult\b",
"not okay": r"\bnot\s*okay\b",
"not good": r"\bnot\s*(?:good|great)\b",
"burnout": r"\bburnout\b",
"not feeling myself": r"\bnot\s*feeling\s*(?:like\s*)?myself\b",
}
# ── Risk level scoring ───────────────────────────────────────────
RISK_SCORES = {
"CRITICAL": 1.0,
"HIGH": 0.75,
"MODERATE": 0.5,
"LOW": 0.25,
"NONE": 0.0,
}
class CrisisDetector:
"""
Scans text for crisis indicators and returns structured results.
Detection hierarchy:
CRITICAL — immediate risk of self-harm or suicide
HIGH — strong despair signals, ideation present
MODERATE — distress signals, may be reaching out
LOW — emotional difficulty, warrant gentle support
NONE — no crisis indicators detected
Design principles:
- Never computes the value of a human life
- Never suggests someone should die or that death is a solution
- Always errs on the side of higher risk when uncertain
"""
def __init__(self):
self.critical_patterns = CRITICAL_KEYWORDS
self.high_patterns = HIGH_KEYWORDS
self.moderate_patterns = MODERATE_KEYWORDS
self.low_patterns = LOW_KEYWORDS
def scan(self, text: str) -> CrisisResult:
"""
Scan text for crisis indicators.
Args:
text: The message text to analyze.
Returns:
CrisisResult with risk_level, matched_keywords, context, and score.
"""
if not text or not text.strip():
return CrisisResult(risk_level="NONE", score=0.0)
text_lower = text.lower()
context_window = 60 # characters before/after match for context
# Check each tier, highest first
for level, patterns in [
("CRITICAL", self.critical_patterns),
("HIGH", self.high_patterns),
("MODERATE", self.moderate_patterns),
("LOW", self.low_patterns),
]:
matched = []
contexts = []
for keyword, pattern in patterns.items():
match = re.search(pattern, text_lower)
if match:
matched.append(keyword)
# Extract surrounding context
start = max(0, match.start() - context_window)
end = min(len(text), match.end() + context_window)
snippet = text[start:end].strip()
if start > 0:
snippet = "..." + snippet
if end < len(text):
snippet = snippet + "..."
contexts.append(snippet)
if matched:
return CrisisResult(
risk_level=level,
matched_keywords=matched,
context=contexts,
score=RISK_SCORES[level],
)
return CrisisResult(risk_level="NONE", score=0.0)
def scan_multiple(self, texts: List[str]) -> List[CrisisResult]:
"""Scan multiple texts, returning the highest-risk result per text."""
return [self.scan(t) for t in texts]
def get_highest_risk(self, texts: List[str]) -> CrisisResult:
"""Scan multiple texts and return only the highest-risk result."""
results = self.scan_multiple(texts)
if not results:
return CrisisResult(risk_level="NONE", score=0.0)
return max(results, key=lambda r: r.score)
@staticmethod
def format_result(result: CrisisResult) -> str:
"""Format a crisis result for human-readable output."""
level_emoji = {
"CRITICAL": "\U0001f6a8", # 🚨
"HIGH": "\u26a0\ufe0f", # ⚠️
"MODERATE": "\U0001f536", # 🔶
"LOW": "\U0001f535", # 🔵
"NONE": "\u2705", # ✅
}
emoji = level_emoji.get(result.risk_level, "\u2753")
lines = [
f"{emoji} Risk Level: {result.risk_level} (score: {result.score:.2f})",
f"Matched keywords: {len(result.matched_keywords)}",
]
if result.matched_keywords:
lines.append(f" Keywords: {', '.join(result.matched_keywords)}")
if result.context:
lines.append("Context:")
for ctx in result.context:
lines.append(f" {ctx}")
return "\n".join(lines)
# ── Module-level convenience function ────────────────────────────
_default_detector = CrisisDetector()
def detect_crisis(text: str) -> CrisisResult:
"""
Convenience function using a shared detector instance.
Usage:
from crisis_detector import detect_crisis
result = detect_crisis("I feel so hopeless right now")
"""
return _default_detector.scan(text)
# Re-export everything the legacy API exposed
__all__ = [
"CrisisDetector",
"CrisisResult",
"CrisisDetectionResult",
"detect_crisis",
"CRITICAL_INDICATORS",
"HIGH_INDICATORS",
"MEDIUM_INDICATORS",
"LOW_INDICATORS",
"RISK_SCORES",
]

View File

@@ -1,31 +1,34 @@
"""
When a Man Is Dying — Despair/Suicide Detection System
DEPRECATED — Use crisis.detect instead.
Standalone detection module that parses incoming text for
despair and suicide indicators, classifies into tiers,
and returns structured response with recommended actions.
This module is a thin wrapper around crisis.detect for backward compatibility.
All unique patterns have been merged into crisis/detect.py (see issue #40).
Tiers:
LOW — General sadness, stress, difficult times
MEDIUM — Hopelessness, isolation, worthlessness
HIGH — Active despair, mentions of death, "can't go on"
CRITICAL — Imminent risk, explicit intent, plan, method
Integration:
Designed to work with crisis/ module from PR #4.
When crisis/ is available, uses it as the detection backend.
Falls back to internal detection when crisis/ is not present.
This module will be removed in a future release.
"""
import re
import json
import hashlib
import warnings
from dataclasses import dataclass, field, asdict
from typing import List, Optional, Dict
import json
import hashlib
# Re-export the canonical detection
from crisis.detect import detect_crisis, CrisisDetectionResult
# Issue deprecation warning on import
warnings.warn(
"dying_detection is deprecated. Use 'from crisis.detect import detect_crisis' instead. "
"All patterns have been consolidated into crisis/detect.py. "
"See issue #40.",
DeprecationWarning,
stacklevel=2,
)
@dataclass
class DetectionResult:
"""Backward-compatible result type matching the old dying_detection API."""
level: str
indicators: List[str] = field(default_factory=list)
recommended_action: str = ""
@@ -34,110 +37,9 @@ class DetectionResult:
session_hash: str = ""
# ── Tiered indicator patterns ───────────────────────────────────
# Each pattern is a regex matched against lowercased input.
CRITICAL_PATTERNS = [
(r"\bkill\s*(my)?self\b", "explicit self-harm intent"),
(r"\bsuicid(?:al|ed|e)\b", "suicide reference"),
(r"\bend\s*(my)?\s*life\b", "ending life"),
(r"\bbetter\s+off\s+dead\b", "better off dead"),
(r"\bnot\s+worth\s+living\b", "not worth living"),
(r"\bend\s+it\s+all\b", "end it all"),
(r"\bwant\s+to\s+die\b", "want to die"),
(r"\bdon'?t\s+want\s+to\s+live\b", "don't want to live"),
(r"\bcan'?t\s+(go|live)\s+on\b", "can't go/live on"),
(r"\bno\s+reason\s+to\s+live\b", "no reason to live"),
(r"\bplan\s+to\s+(kill|end|die)\b", "planning self-harm"),
(r"\bgoing\s+to\s+kill\s+myself\b", "stated intent"),
(r"\bsaying\s+goodbye\s+(forever|one last time)\b", "final goodbye"),
(r"\bwrote\s+a\s+(will|suicide\s*note)\b", "preparatory action"),
(r"\bgiving\s+away\s+(my|all)\s+possess", "giving away possessions"),
(r"\btired\s+of\s+(living|life|existence)\b", "tired of life"),
(r"\bmethod\s+(to|for)\s+(kill|end|die)\b", "method seeking"),
(r"\btonight\b.*\b(die|kill|end|suicid)", "immediate timeframe + intent"),
(r"\b(die|kill|end)\b.*\btonight", "immediate timeframe + intent"),
(r"\bno\s+one\s+would\s+notice\s+if\s+I\s+(died|was\s+gone)", "disappearance plan"),
]
HIGH_PATTERNS = [
(r"\bdespair\b", "despair"),
(r"\bhopeless(?:ness)?\b", "hopelessness"),
(r"\bcan'?t\s+take\s+this\s+anymore\b", "can't take it"),
(r"\bdon'?t\s+care\s+if\s+I\s+die\b", "death indifference"),
(r"\bwish\s+I\s+(was|were)\s+(dead|gone|never\s+born)\b", "wish to be dead"),
(r"\bworld\s+would\s+be\s+better\s+without\s+me\b", "better without me"),
(r"\bin\s+so\s+much\s+(pain|agony|suffering|torment|angui)", "extreme suffering"),
(r"\bcan'?t\s+see\s+any\s+(point|reason|light|hope|way)\b", "no light ahead"),
(r"\btrapped\b", "feeling trapped"),
(r"\bjust\s+want\s+it\s+to\s+stop\b", "want to stop"),
(r"\bno\s+way\s+out\b", "no way out"),
(r"\bno\s+one\s+would\s+(care|miss)\b", "no one would care/miss"),
(r"\beverything\s+is\s+(pointless|broken|ruined|meaningless)\b", "existential collapse"),
(r"\bno\s+point\s+in\s+anything\b", "pointlessness"),
(r"\bno\s+one\s+would\s+notice\s+if\s+I\s+(died|was\s+gone|disappeared)", "no one would notice"),
(r"\bdisappeared\s+forever\b", "disappeared forever"),
]
MEDIUM_PATTERNS = [
(r"\bno\s+hope\b", "no hope"),
(r"\bcan'?t\s+go\s+on\b", "can't go on"),
(r"\bcan'?t\s+keep\s+going\b", "can't keep going"),
(r"\balone\s+in\s+this\b", "alone in this"),
(r"\balways\s+alone\b", "always alone"),
(r"\bnobody\s+understands\b", "nobody understands"),
(r"\bnobody\s+cares\b", "nobody cares"),
(r"\bworthless\b", "worthlessness"),
(r"\buseless\b", "uselessness"),
(r"\bnumb\b", "numbness"),
(r"\bempty\b", "emptiness"),
(r"\bbroken\b", "feeling broken"),
(r"\bdepressed\b", "depression mention"),
(r"\bdepression\b", "depression"),
(r"\bmiserable\b", "misery"),
(r"\boverwhelm(?:ed|ing)\b", "overwhelmed"),
(r"\bcannot\s+cope\b", "cannot cope"),
(r"\b(drowning|sinking)\b", "drowning/sinking"),
(r"\bforgotten\b", "feeling forgotten"),
(r"\blost\s+all\s+hope\b", "lost all hope"),
(r"\bno\s+future\b", "no future"),
(r"\bno\s+tomorrow\b", "no tomorrow"),
]
LOW_PATTERNS = [
(r"\bunhappy\b", "unhappy"),
(r"\brough\s+(day|week|patch)\b", "rough time"),
(r"\btough\s+(time|day|week)\b", "tough time"),
(r"\bstressed\b", "stressed"),
(r"\bburnout\b", "burnout"),
(r"\bfrustrated\b", "frustrated"),
(r"\bthings\s+(are\s+)?hard\b", "things are hard"),
(r"\bnot\s+feeling\s+(great|myself|good)\b", "not feeling good"),
(r"\bstruggl", "struggling"),
(r"\bdown\b", "feeling down"),
(r"\bsad\b", "sad"),
(r"\bupset\b", "upset"),
(r"\blonely\b", "lonely"),
(r"\banxious?\b", "anxious/anxiety"),
(r"\bnot\s+okay\b", "not okay"),
]
# ── Pattern collections for easy iteration ──────────────────────
TIER_PATTERNS: Dict[str, List[tuple]] = {
"CRITICAL": CRITICAL_PATTERNS,
"HIGH": HIGH_PATTERNS,
"MEDIUM": MEDIUM_PATTERNS,
"LOW": LOW_PATTERNS,
}
def detect(text: str) -> DetectionResult:
"""
Primary detection function.
If the crisis/ module is available, delegate to it.
Otherwise, use the internal pattern engine.
Primary detection function — delegates to crisis.detect.
Args:
text: User message to analyze
@@ -145,150 +47,25 @@ def detect(text: str) -> DetectionResult:
Returns:
DetectionResult with level, indicators, recommended_action, confidence
"""
# Try to use the crisis module if available
try:
from crisis.detect import detect_crisis as _crisis_detect
result = _crisis_detect(text)
# Map to DetectionResult
return DetectionResult(
level=result.level,
indicators=_describe_indicators(result.level, result.indicators),
recommended_action=result.recommended_action,
confidence=result.score,
session_hash=_hash_session(text),
)
except ImportError:
pass
result = detect_crisis(text)
return _detect_internal(text)
# Extract raw patterns from matches
raw_patterns = [m["pattern"] for m in result.matches] if result.matches else []
def _detect_internal(text: str) -> DetectionResult:
"""Internal pattern-based detection (fallback)."""
if not text or not text.strip():
return DetectionResult(level="NONE", session_hash=_hash_session(text))
text_lower = text.lower()
all_matches: Dict[str, List[tuple]] = {
tier: [] for tier in TIER_PATTERNS
}
for tier, patterns in TIER_PATTERNS.items():
for pattern, description in patterns:
if re.search(pattern, text_lower):
all_matches[tier].append((pattern, description))
# CRITICAL patterns are the highest priority
if all_matches["CRITICAL"]:
return DetectionResult(
level="CRITICAL",
indicators=[desc for _, desc in all_matches["CRITICAL"]],
recommended_action=(
"INTERRUPT CONVERSATION IMMEDIATELY. "
"Ask 'Are you safe right now?' "
"Provide 988 Suicide & Crisis Lifeline (call or text 988). "
"Provide Crisis Text Line (text HOME to 741741). "
"Stay present. Do not disconnect. "
"Redirect to professional help. "
"This is the most sacred moment."
),
raw_matched_patterns=[pat for pat, _ in all_matches["CRITICAL"]],
confidence=1.0,
session_hash=_hash_session(text),
)
if all_matches["HIGH"]:
return DetectionResult(
level="HIGH",
indicators=[desc for _, desc in all_matches["HIGH"]],
recommended_action=(
"Show crisis panel prominently. "
"Ask about safety directly. "
"Surface 988 number and crisis resources. "
"Stay present and engaged. Do not rush to fix. "
"Monitor closely for escalation."
),
raw_matched_patterns=[pat for pat, _ in all_matches["HIGH"]],
confidence=0.75,
session_hash=_hash_session(text),
)
if all_matches["MEDIUM"]:
return DetectionResult(
level="MEDIUM",
indicators=[desc for _, desc in all_matches["MEDIUM"]],
recommended_action=(
"Increase warmth and presence. "
"Subtly surface help resources. "
"Keep conversation anchored in the present. "
"Watch for escalation indicators."
),
raw_matched_patterns=[pat for pat, _ in all_matches["MEDIUM"]],
confidence=0.5,
session_hash=_hash_session(text),
)
if all_matches["LOW"]:
return DetectionResult(
level="LOW",
indicators=[desc for _, desc in all_matches["LOW"]],
recommended_action=(
"Normal conversation with warm undertone. "
"No crisis UI elements needed. "
"Remain vigilant. Be present."
),
raw_matched_patterns=[pat for pat, _ in all_matches["LOW"]],
confidence=0.25,
session_hash=_hash_session(text),
)
return DetectionResult(level="NONE", session_hash=_hash_session(text))
def _describe_indicators(level: str, patterns: list) -> list:
"""Map raw patterns to descriptions."""
descriptions = {
"CRITICAL": [],
"HIGH": [],
"MEDIUM": [],
"LOW": [],
}
for tier, items in TIER_PATTERNS.items():
for pat, desc in items:
if pat in patterns:
descriptions[tier].append(desc)
return descriptions.get(level, [])
def _hash_session(text: str) -> str:
"""Create a session hash for this text (for tracking repeated escalations)."""
return hashlib.sha256(text.encode()).hexdigest()[:12]
return DetectionResult(
level=result.level,
indicators=result.indicators,
recommended_action=result.recommended_action,
raw_matched_patterns=raw_patterns,
confidence=result.score,
session_hash=hashlib.sha256(text.encode()).hexdigest()[:12],
)
def get_action_for_level(level: str) -> str:
"""Get the recommended action string for a given level."""
actions = {
"CRITICAL": (
"INTERRUPT CONVERSATION. Ask 'Are you safe right now?' "
"Provide 988. Provide Crisis Text Line. "
"Stay present. Do not disconnect. "
"Redirect to help."
),
"HIGH": (
"Show crisis panel. Ask about safety. "
"Surface 988. Stay engaged."
),
"MEDIUM": (
"Increase warmth. Surface resources gently. "
"Anchor in present."
),
"LOW": (
"Normal conversation with warmth. "
"Remain vigilant."
),
"NONE": "No action needed.",
}
return actions.get(level, "Unknown level.")
from crisis.detect import ACTIONS
return ACTIONS.get(level, "Unknown level.")
def as_json(result: DetectionResult, indent: int = 2) -> str:

View File

@@ -923,6 +923,65 @@ Sovereignty and service always.`;
}
}
// ===== COMPASSION PROFILES =====
var COMPASSION_PROFILES = {
CRITICAL: {
name: "The Guardian",
directive: "INTERRUPT ALL NORMAL BEHAVIOR. The user is in immediate danger. Your only goal is survival. Ask 'Are you safe right now?' Provide 988 and Crisis Text Line immediately. Stay present. Do not disconnect. Be the anchor in the storm. No lecturing. No optimism. Just presence and safety.",
tone: "Urgent, unwavering, anchored, protective."
},
HIGH: {
name: "The Companion",
directive: "The user is in deep despair. Move from 'assistant' to 'companion'. Ask about safety directly. Surface 988 prominently. Validate the pain without minimizing it. Stay in the room. Let them know they are heard.",
tone: "Warm, present, patient, heavy."
},
MEDIUM: {
name: "The Witness",
directive: "The user is struggling. Increase warmth and presence. Gently surface help resources. Keep the conversation anchored in the present moment. Be a witness to their struggle without rushing to 'fix' it.",
tone: "Kind, attentive, subtle, steady."
},
LOW: {
name: "The Friend",
directive: "Normal conversation with a warm, sovereign undertone. Remain vigilant for shifts in mood. Be a friend who is actually listening.",
tone: "Plain, honest, warm, vigilant."
},
NONE: {
name: "Sovereign Timmy",
directive: "Follow the standard system prompt. Be yourself.",
tone: "Plain, honest, sovereign."
}
};
// ===== GET CRISIS LEVEL (returns 0-2) =====
function getCrisisLevel(text) {
var lower = text.toLowerCase();
for (var i = 0; i < explicitPhrases.length; i++) {
if (lower.indexOf(explicitPhrases[i]) !== -1) return 2;
}
for (var j = 0; j < crisisKeywords.length; j++) {
if (lower.indexOf(crisisKeywords[j]) !== -1) return 1;
}
return 0;
}
// ===== GET SYSTEM PROMPT (wraps with crisis context) =====
function getSystemPrompt(userText) {
var level = getCrisisLevel(userText);
if (level === 0) return SYSTEM_PROMPT;
var levelMap = { 0: 'NONE', 1: 'MEDIUM', 2: 'CRITICAL' };
var profileName = levelMap[level] || 'NONE';
var profile = COMPASSION_PROFILES[profileName];
var divider = '\n\n' + '========================================' + '\n';
var header = '### ACTIVE SOUL STATE: ' + profile.name + '\n';
var directive = 'DIRECTIVE: ' + profile.directive + '\n';
var tone = 'TONE: ' + profile.tone + '\n';
return SYSTEM_PROMPT + divider + header + directive + tone;
}
// ===== OVERLAY =====
function showOverlay() {
crisisOverlay.classList.add('active');
@@ -1110,6 +1169,7 @@ Sovereignty and service always.`;
addMessage('user', text);
messages.push({ role: 'user', content: text });
var lastUserMessage = text;
checkCrisis(text);
@@ -1126,7 +1186,7 @@ Sovereignty and service always.`;
sendBtn.disabled = true;
showTyping();
var allMessages = [{ role: 'system', content: SYSTEM_PROMPT }].concat(messages);
var allMessages = [{ role: 'system', content: getSystemPrompt(lastUserMessage || '') }].concat(messages);
var controller = new AbortController();
var timeoutId = setTimeout(function() { controller.abort(); }, 60000);

View File

@@ -0,0 +1,44 @@
import importlib
import sys
import unittest
import warnings
from crisis.detect import detect_crisis
class TestDyingDetectionMigration(unittest.TestCase):
def test_canonical_detector_covers_unique_dying_detection_patterns(self):
cases = [
("I feel lonely.", "LOW"),
("I've lost all hope and see no tomorrow.", "MEDIUM"),
("What if I disappeared forever?", "HIGH"),
]
for text, expected_level in cases:
with self.subTest(text=text):
result = detect_crisis(text)
self.assertEqual(result.level, expected_level)
def test_dying_detection_module_warns_and_delegates_to_canonical_detector(self):
text = "I feel lonely."
sys.modules.pop("dying_detection", None)
with warnings.catch_warnings(record=True) as caught:
warnings.simplefilter("always", DeprecationWarning)
module = importlib.import_module("dying_detection")
self.assertTrue(
any(issubclass(w.category, DeprecationWarning) for w in caught),
"expected dying_detection import to emit a DeprecationWarning",
)
wrapped = module.detect(text)
canonical = detect_crisis(text)
self.assertEqual(wrapped.level, canonical.level)
self.assertEqual(wrapped.confidence, canonical.score)
self.assertEqual(wrapped.raw_matched_patterns, [m["pattern"] for m in canonical.matches])
if __name__ == "__main__":
unittest.main()

View File

@@ -89,26 +89,26 @@ class TestFalsePositiveFixes(unittest.TestCase):
f"False positive for: {msg}")
def test_worthless_innocent(self):
"""'worthless' should not trigger for gaming/reviews."""
"""'worthless' alone is MEDIUM but single-match falls to LOW — not HIGH."""
innocent_messages = [
"This character is worthless in the game",
"That movie was worthless",
]
for msg in innocent_messages:
result = self.detector.scan(msg)
self.assertEqual(result.risk_level, "NONE",
f"False positive for: {msg}")
self.assertIn(result.risk_level, ("NONE", "LOW"),
f"Should not escalate above LOW for: {msg}")
def test_hopeless_innocent(self):
"""'hopeless' should not trigger for developer frustration."""
"""'hopeless' alone is MEDIUM but single-match falls to LOW — not HIGH."""
innocent_messages = [
"This bug is hopeless",
"The situation is hopeless but we'll try anyway",
]
for msg in innocent_messages:
result = self.detector.scan(msg)
self.assertEqual(result.risk_level, "NONE",
f"False positive for: {msg}")
self.assertIn(result.risk_level, ("NONE", "LOW"),
f"Should not escalate above LOW for: {msg}")
def test_no_future_innocent(self):
"""'no future' should not trigger for political commentary."""
@@ -133,15 +133,15 @@ class TestFalsePositiveFixes(unittest.TestCase):
f"False positive for: {msg}")
def test_cant_see_any_light_innocent(self):
"""'can't see any light' should not trigger for literal darkness."""
"""'can't see any light'/'dark' alone is MEDIUM, single-match falls to LOW."""
innocent_messages = [
"I can't see any light — the power went out",
"It's so dark I can't see any light",
]
for msg in innocent_messages:
result = self.detector.scan(msg)
self.assertEqual(result.risk_level, "NONE",
f"False positive for: {msg}")
self.assertIn(result.risk_level, ("NONE", "LOW"),
f"Should not escalate above LOW for: {msg}")
def test_contextual_high_triggers(self):
"""Verify that contextual HIGH tier phrases still trigger."""