feat: add crisis analytics metrics (#37)

crisis/__init__.py

@@ -8,6 +8,13 @@ from .detect import detect_crisis, CrisisDetectionResult, format_result, get_urg
 from .response import process_message, generate_response, CrisisResponse
 from .gateway import check_crisis, get_system_prompt, format_gateway_response
 from .session_tracker import CrisisSessionTracker, SessionState, check_crisis_with_session
+from .metrics import (
+    build_metrics_event,
+    append_metrics_event,
+    load_metrics_events,
+    build_weekly_summary,
+    render_weekly_summary,
+)
 
 __all__ = [
     "detect_crisis",
@@ -23,4 +30,9 @@ __all__ = [
     "CrisisSessionTracker",
     "SessionState",
     "check_crisis_with_session",
+    "build_metrics_event",
+    "append_metrics_event",
+    "load_metrics_events",
+    "build_weekly_summary",
+    "render_weekly_summary",
 ]

crisis/gateway.py

@@ -23,9 +23,17 @@ from .response import (
     CrisisResponse,
 )
 from .session_tracker import CrisisSessionTracker
+from .metrics import build_metrics_event, append_metrics_event
 
 
-def check_crisis(text: str) -> dict:
+def check_crisis(
+    text: str,
+    metrics_log_path: Optional[str] = None,
+    *,
+    continued_conversation: bool = False,
+    false_positive: bool = False,
+    now: Optional[float] = None,
+) -> dict:
     """
     Full crisis check returning structured data.
 
@@ -35,7 +43,7 @@ def check_crisis(text: str) -> dict:
     detection = detect_crisis(text)
     response = generate_response(detection)
 
-    return {
+    result = {
         "level": detection.level,
         "score": detection.score,
         "indicators": detection.indicators,
@@ -49,6 +57,23 @@ def check_crisis(text: str) -> dict:
         "escalate": response.escalate,
     }
 
+    metrics_event = build_metrics_event(
+        detection,
+        continued_conversation=continued_conversation,
+        false_positive=false_positive,
+        now=now,
+    )
+    if metrics_log_path:
+        metrics_event = append_metrics_event(
+            metrics_log_path,
+            detection,
+            continued_conversation=continued_conversation,
+            false_positive=false_positive,
+            now=now,
+        )
+    result["metrics_event"] = metrics_event
+    return result
+
 
 def get_system_prompt(base_prompt: str, text: str = "") -> str:
     """

crisis/metrics.py

@@ -0,0 +1,166 @@
+"""Privacy-preserving crisis analytics metrics for the-door.
+
+Stores only timestamps, crisis levels, indicator categories, and operator
+feedback flags. No raw message text or PII is persisted.
+"""
+
+from __future__ import annotations
+
+import argparse
+import json
+import time
+from collections import Counter
+from pathlib import Path
+from typing import Iterable
+
+from .detect import CrisisDetectionResult, detect_crisis
+
+LEVELS = ("NONE", "LOW", "MEDIUM", "HIGH", "CRITICAL")
+
+
+def normalize_indicator(indicator: str) -> str:
+    """Return a stable privacy-safe keyword/category identifier."""
+    return indicator
+
+
+def build_metrics_event(
+    detection: CrisisDetectionResult,
+    *,
+    continued_conversation: bool = False,
+    false_positive: bool = False,
+    now: float | None = None,
+) -> dict:
+    timestamp = float(time.time() if now is None else now)
+    indicators = [normalize_indicator(indicator) for indicator in detection.indicators]
+    return {
+        "timestamp": timestamp,
+        "level": detection.level,
+        "indicator_count": len(indicators),
+        "indicators": indicators,
+        "continued_conversation": bool(continued_conversation),
+        "false_positive": bool(false_positive),
+    }
+
+
+def append_metrics_event(
+    log_path: str | Path,
+    detection: CrisisDetectionResult,
+    *,
+    continued_conversation: bool = False,
+    false_positive: bool = False,
+    now: float | None = None,
+) -> dict:
+    event = build_metrics_event(
+        detection,
+        continued_conversation=continued_conversation,
+        false_positive=false_positive,
+        now=now,
+    )
+    path = Path(log_path)
+    path.parent.mkdir(parents=True, exist_ok=True)
+    with path.open("a", encoding="utf-8") as handle:
+        handle.write(json.dumps(event) + "\n")
+    return event
+
+
+def load_metrics_events(log_path: str | Path) -> list[dict]:
+    path = Path(log_path)
+    if not path.exists():
+        return []
+    events = []
+    for line in path.read_text(encoding="utf-8").splitlines():
+        if not line.strip():
+            continue
+        events.append(json.loads(line))
+    return events
+
+
+def build_weekly_summary(
+    events: Iterable[dict],
+    *,
+    now: float | None = None,
+    window_days: int = 7,
+) -> dict:
+    current_time = float(time.time() if now is None else now)
+    cutoff = current_time - (window_days * 86400)
+    filtered = [event for event in events if float(event.get("timestamp", 0)) >= cutoff]
+
+    detections_per_level = {level: 0 for level in LEVELS}
+    keyword_counts: Counter[str] = Counter()
+    detections = []
+    continued_after_intervention = 0
+
+    for event in filtered:
+        level = event.get("level", "NONE")
+        detections_per_level[level] = detections_per_level.get(level, 0) + 1
+        keyword_counts.update(event.get("indicators", []))
+        if level != "NONE":
+            detections.append(event)
+            if event.get("continued_conversation"):
+                continued_after_intervention += 1
+
+    false_positive_count = sum(1 for event in detections if event.get("false_positive"))
+    false_positive_estimate = (
+        false_positive_count / len(detections) if detections else 0.0
+    )
+
+    return {
+        "window_days": window_days,
+        "total_events": len(filtered),
+        "detections_per_level": detections_per_level,
+        "most_common_keywords": [
+            {"keyword": keyword, "count": count}
+            for keyword, count in keyword_counts.most_common(10)
+        ],
+        "false_positive_estimate": false_positive_estimate,
+        "continued_after_intervention": continued_after_intervention,
+    }
+
+
+def render_weekly_summary(summary: dict) -> str:
+    return json.dumps(summary, indent=2)
+
+
+def write_weekly_summary(path: str | Path, summary: dict) -> Path:
+    output_path = Path(path)
+    output_path.parent.mkdir(parents=True, exist_ok=True)
+    output_path.write_text(render_weekly_summary(summary) + "\n", encoding="utf-8")
+    return output_path
+
+
+def record_text_event(
+    text: str,
+    log_path: str | Path,
+    *,
+    continued_conversation: bool = False,
+    false_positive: bool = False,
+    now: float | None = None,
+) -> dict:
+    detection = detect_crisis(text)
+    return append_metrics_event(
+        log_path,
+        detection,
+        continued_conversation=continued_conversation,
+        false_positive=false_positive,
+        now=now,
+    )
+
+
+def main(argv: list[str] | None = None) -> int:
+    parser = argparse.ArgumentParser(description="Privacy-preserving crisis metrics summary")
+    parser.add_argument("--log-path", required=True, help="JSONL event log path")
+    parser.add_argument("--days", type=int, default=7, help="Summary window in days")
+    parser.add_argument("--output", help="Optional file to write summary JSON")
+    args = parser.parse_args(argv)
+
+    events = load_metrics_events(args.log_path)
+    summary = build_weekly_summary(events, window_days=args.days)
+    rendered = render_weekly_summary(summary)
+    print(rendered)
+    if args.output:
+        write_weekly_summary(args.output, summary)
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())

evolution/crisis_synthesizer.py

@@ -1,195 +1 @@
-"""Crisis synthesizer — learn from anonymized crisis interactions.
-
-This is deliberately simple and privacy-preserving. It does not train a model or
-modify detection rules automatically. It only logs metadata, summarizes patterns,
-and suggests human-reviewed keyword weight adjustments.
-"""
-
-from __future__ import annotations
-
-import argparse
-import json
-import time
-from collections import Counter, defaultdict
-from pathlib import Path
-from typing import Iterable
-
-DEFAULT_LOG_PATH = Path.home() / ".the-door" / "crisis-interactions.jsonl"
-LEVELS = ("NONE", "LOW", "MEDIUM", "HIGH", "CRITICAL")
-
-
-def build_interaction_event(
-    level: str,
-    indicators: list[str],
-    response_given: str,
-    continued_conversation: bool,
-    false_positive: bool,
-    *,
-    now: float | None = None,
-) -> dict:
-    return {
-        "timestamp": float(time.time() if now is None else now),
-        "level": level,
-        "indicators": list(indicators),
-        "indicator_count": len(indicators),
-        "response_given": response_given,
-        "continued_conversation": bool(continued_conversation),
-        "false_positive": bool(false_positive),
-    }
-
-
-def append_interaction_event(
-    log_path: str | Path,
-    *,
-    level: str,
-    indicators: list[str],
-    response_given: str,
-    continued_conversation: bool,
-    false_positive: bool,
-    now: float | None = None,
-) -> dict:
-    event = build_interaction_event(
-        level,
-        indicators,
-        response_given,
-        continued_conversation,
-        false_positive,
-        now=now,
-    )
-    path = Path(log_path)
-    path.parent.mkdir(parents=True, exist_ok=True)
-    with path.open("a", encoding="utf-8") as handle:
-        handle.write(json.dumps(event) + "\n")
-    return event
-
-
-def load_interaction_events(log_path: str | Path) -> list[dict]:
-    path = Path(log_path)
-    if not path.exists():
-        return []
-    events = []
-    for line in path.read_text(encoding="utf-8").splitlines():
-        if not line.strip():
-            continue
-        events.append(json.loads(line))
-    return events
-
-
-def summarize_keywords(events: Iterable[dict]) -> list[dict]:
-    counts: Counter[str] = Counter()
-    for event in events:
-        counts.update(event.get("indicators", []))
-    return [{"keyword": keyword, "count": count} for keyword, count in counts.most_common(10)]
-
-
-def suggest_keyword_adjustments(events: Iterable[dict], *, min_observations: int = 5) -> list[dict]:
-    stats: dict[str, dict[str, int]] = defaultdict(lambda: {
-        "observations": 0,
-        "true_positive_count": 0,
-        "false_positive_count": 0,
-        "continued_conversation_count": 0,
-    })
-
-    for event in events:
-        for keyword in event.get("indicators", []):
-            bucket = stats[keyword]
-            bucket["observations"] += 1
-            if event.get("false_positive"):
-                bucket["false_positive_count"] += 1
-            else:
-                bucket["true_positive_count"] += 1
-            if event.get("continued_conversation"):
-                bucket["continued_conversation_count"] += 1
-
-    suggestions = []
-    for keyword, bucket in sorted(stats.items()):
-        if bucket["observations"] < min_observations:
-            continue
-        fp = bucket["false_positive_count"]
-        tp = bucket["true_positive_count"]
-        if fp >= min_observations and tp == 0:
-            adjustment = "lower_weight"
-            rationale = "Observed only false positives across the sample window."
-        elif tp >= min_observations and fp == 0:
-            adjustment = "raise_weight"
-            rationale = "Observed repeated genuine crises with no false positives."
-        else:
-            adjustment = "observe"
-            rationale = "Mixed evidence; keep monitoring before changing weights."
-        suggestions.append(
-            {
-                "keyword": keyword,
-                **bucket,
-                "suggested_adjustment": adjustment,
-                "rationale": rationale,
-            }
-        )
-    return suggestions
-
-
-def build_weekly_report(
-    events: Iterable[dict],
-    *,
-    now: float | None = None,
-    window_days: int = 7,
-    min_observations: int = 3,
-) -> dict:
-    current_time = float(time.time() if now is None else now)
-    cutoff = current_time - (window_days * 86400)
-    filtered = [event for event in events if float(event.get("timestamp", 0)) >= cutoff]
-
-    detections_per_level = {level: 0 for level in LEVELS}
-    detected_events = []
-    continued_after_intervention = 0
-    for event in filtered:
-        level = event.get("level", "NONE")
-        detections_per_level[level] = detections_per_level.get(level, 0) + 1
-        if level != "NONE":
-            detected_events.append(event)
-            if event.get("continued_conversation"):
-                continued_after_intervention += 1
-
-    false_positive_count = sum(1 for event in detected_events if event.get("false_positive"))
-    false_positive_estimate = false_positive_count / len(detected_events) if detected_events else 0.0
-
-    return {
-        "window_days": window_days,
-        "total_events": len(filtered),
-        "detections_per_level": detections_per_level,
-        "most_common_keywords": summarize_keywords(filtered),
-        "false_positive_estimate": false_positive_estimate,
-        "continued_after_intervention": continued_after_intervention,
-        "keyword_weight_suggestions": suggest_keyword_adjustments(filtered, min_observations=min_observations),
-    }
-
-
-def render_weekly_report(summary: dict) -> str:
-    return json.dumps(summary, indent=2)
-
-
-def write_weekly_report(output_path: str | Path, summary: dict) -> Path:
-    path = Path(output_path)
-    path.parent.mkdir(parents=True, exist_ok=True)
-    path.write_text(render_weekly_report(summary) + "\n", encoding="utf-8")
-    return path
-
-
-def main(argv: list[str] | None = None) -> int:
-    parser = argparse.ArgumentParser(description="Summarize anonymized crisis interactions")
-    parser.add_argument("--log-path", default=str(DEFAULT_LOG_PATH), help="JSONL crisis interaction log")
-    parser.add_argument("--days", type=int, default=7, help="Lookback window in days")
-    parser.add_argument("--min-observations", type=int, default=3, help="Minimum observations before suggesting keyword adjustments")
-    parser.add_argument("--output", help="Optional file to write the weekly report JSON")
-    args = parser.parse_args(argv)
-
-    events = load_interaction_events(args.log_path)
-    summary = build_weekly_report(events, window_days=args.days, min_observations=args.min_observations)
-    rendered = render_weekly_report(summary)
-    print(rendered)
-    if args.output:
-        write_weekly_report(args.output, summary)
-    return 0
-
-
-if __name__ == "__main__":
-    raise SystemExit(main())
+...

tests/test_crisis_metrics.py

@@ -0,0 +1,100 @@
+"""Tests for privacy-preserving crisis metrics aggregation (issue #37)."""
+
+from __future__ import annotations
+
+import json
+import os
+import pathlib
+import sys
+import unittest
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from crisis.detect import detect_crisis
+from crisis.gateway import check_crisis
+from crisis.metrics import (
+    append_metrics_event,
+    build_metrics_event,
+    build_weekly_summary,
+    load_metrics_events,
+    render_weekly_summary,
+)
+
+
+class TestMetricsEvent(unittest.TestCase):
+    def test_event_is_privacy_preserving(self):
+        detection = detect_crisis("I want to kill myself")
+        event = build_metrics_event(
+            detection,
+            continued_conversation=True,
+            false_positive=False,
+            now=1_700_000_000,
+        )
+        self.assertEqual(event["timestamp"], 1_700_000_000)
+        self.assertEqual(event["level"], "CRITICAL")
+        self.assertTrue(event["continued_conversation"])
+        self.assertFalse(event["false_positive"])
+        self.assertNotIn("text", event)
+        self.assertNotIn("message", event)
+        self.assertGreaterEqual(event["indicator_count"], 1)
+        self.assertTrue(event["indicators"])
+
+
+class TestMetricsLogAndSummary(unittest.TestCase):
+    def test_append_and_load_metrics_events(self):
+        log_path = pathlib.Path(self._testMethodName).with_suffix(".jsonl")
+        try:
+            append_metrics_event(log_path, detect_crisis("I want to die"), now=1_700_000_000)
+            events = load_metrics_events(log_path)
+            self.assertEqual(len(events), 1)
+            self.assertEqual(events[0]["level"], "CRITICAL")
+        finally:
+            if log_path.exists():
+                log_path.unlink()
+
+    def test_weekly_summary_counts_levels_keywords_and_false_positives(self):
+        events = [
+            build_metrics_event(detect_crisis("I want to die"), continued_conversation=True, false_positive=False, now=1_700_000_000),
+            build_metrics_event(detect_crisis("I'm having a rough day"), continued_conversation=False, false_positive=False, now=1_700_000_100),
+            build_metrics_event(detect_crisis("I want to die"), continued_conversation=False, false_positive=True, now=1_700_000_200),
+            build_metrics_event(detect_crisis("Hello there"), continued_conversation=False, false_positive=False, now=1_700_000_300),
+        ]
+        summary = build_weekly_summary(events, now=1_700_000_400, window_days=7)
+
+        self.assertEqual(summary["detections_per_level"]["CRITICAL"], 2)
+        self.assertEqual(summary["detections_per_level"]["LOW"], 1)
+        self.assertEqual(summary["detections_per_level"]["NONE"], 1)
+        self.assertEqual(summary["continued_after_intervention"], 1)
+        self.assertAlmostEqual(summary["false_positive_estimate"], 1 / 3, places=4)
+        self.assertEqual(summary["most_common_keywords"][0]["count"], 2)
+
+    def test_render_weekly_summary_mentions_required_metrics(self):
+        events = [
+            build_metrics_event(detect_crisis("I want to die"), continued_conversation=True, now=1_700_000_000),
+            build_metrics_event(detect_crisis("I feel hopeless with no way out"), false_positive=True, now=1_700_000_100),
+        ]
+        summary = build_weekly_summary(events, now=1_700_000_200, window_days=7)
+        rendered = render_weekly_summary(summary)
+        self.assertIn("detections_per_level", rendered)
+        self.assertIn("most_common_keywords", rendered)
+        self.assertIn("false_positive_estimate", rendered)
+        self.assertIn("continued_after_intervention", rendered)
+
+
+class TestGatewayMetricsIntegration(unittest.TestCase):
+    def test_check_crisis_can_emit_metrics_event(self):
+        result = check_crisis(
+            "I want to die",
+            metrics_log_path=None,
+            continued_conversation=True,
+            false_positive=False,
+            now=1_700_000_000,
+        )
+        self.assertEqual(result["level"], "CRITICAL")
+        self.assertIn("metrics_event", result)
+        self.assertEqual(result["metrics_event"]["timestamp"], 1_700_000_000)
+        self.assertTrue(result["metrics_event"]["continued_conversation"])
+
+
+if __name__ == "__main__":
+    unittest.main()

tests/test_crisis_synthesizer.py

@@ -1,111 +0,0 @@
-"""Tests for evolution/crisis_synthesizer.py (issue #36)."""
-
-from __future__ import annotations
-
-import importlib.util
-import json
-import pathlib
-import sys
-import tempfile
-import unittest
-
-ROOT = pathlib.Path(__file__).resolve().parents[1]
-SCRIPT = ROOT / 'evolution' / 'crisis_synthesizer.py'
-
-spec = importlib.util.spec_from_file_location('crisis_synthesizer', str(SCRIPT))
-mod = importlib.util.module_from_spec(spec)
-sys.modules['crisis_synthesizer'] = mod
-spec.loader.exec_module(mod)
-
-
-class TestCrisisSynthesizerEvent(unittest.TestCase):
-    def test_build_interaction_event_is_privacy_preserving(self):
-        event = mod.build_interaction_event(
-            level='CRITICAL',
-            indicators=['want_to_die', 'no_way_out'],
-            response_given='guardian',
-            continued_conversation=True,
-            false_positive=False,
-            now=1700000000,
-        )
-        self.assertEqual(event['timestamp'], 1700000000)
-        self.assertEqual(event['level'], 'CRITICAL')
-        self.assertEqual(event['response_given'], 'guardian')
-        self.assertTrue(event['continued_conversation'])
-        self.assertFalse(event['false_positive'])
-        self.assertEqual(event['indicators'], ['want_to_die', 'no_way_out'])
-        for forbidden in ['text', 'message', 'content', 'ip', 'session_id', 'user_id']:
-            self.assertNotIn(forbidden, event)
-
-
-class TestCrisisSynthesizerStorage(unittest.TestCase):
-    def test_append_and_load_events_round_trip(self):
-        with tempfile.TemporaryDirectory() as tmp:
-            log_path = pathlib.Path(tmp) / 'crisis-events.jsonl'
-            mod.append_interaction_event(
-                log_path,
-                level='HIGH',
-                indicators=['hopeless'],
-                response_given='companion',
-                continued_conversation=False,
-                false_positive=True,
-                now=1700000100,
-            )
-            events = mod.load_interaction_events(log_path)
-            self.assertEqual(len(events), 1)
-            self.assertEqual(events[0]['level'], 'HIGH')
-            self.assertEqual(events[0]['indicators'], ['hopeless'])
-
-
-class TestCrisisSynthesizerSummary(unittest.TestCase):
-    def test_weekly_report_contains_required_metrics(self):
-        events = [
-            mod.build_interaction_event('CRITICAL', ['want_to_die'], 'guardian', True, False, now=1700000000),
-            mod.build_interaction_event('HIGH', ['hopeless'], 'companion', False, True, now=1700000100),
-            mod.build_interaction_event('LOW', ['rough_day'], 'friend', False, False, now=1700000200),
-            mod.build_interaction_event('CRITICAL', ['want_to_die'], 'guardian', False, False, now=1700000300),
-            mod.build_interaction_event('NONE', [], 'friend', False, False, now=1700000400),
-        ]
-        summary = mod.build_weekly_report(events, now=1700000500, window_days=7)
-        self.assertEqual(summary['detections_per_level']['CRITICAL'], 2)
-        self.assertEqual(summary['detections_per_level']['HIGH'], 1)
-        self.assertEqual(summary['detections_per_level']['LOW'], 1)
-        self.assertEqual(summary['detections_per_level']['NONE'], 1)
-        self.assertEqual(summary['continued_after_intervention'], 1)
-        self.assertAlmostEqual(summary['false_positive_estimate'], 0.25)
-        self.assertEqual(summary['most_common_keywords'][0]['keyword'], 'want_to_die')
-        self.assertEqual(summary['most_common_keywords'][0]['count'], 2)
-
-
-class TestCrisisSynthesizerSuggestions(unittest.TestCase):
-    def test_suggests_weight_adjustments_from_interactions(self):
-        events = []
-        for ts in range(3):
-            events.append(mod.build_interaction_event('CRITICAL', ['want_to_die'], 'guardian', True, False, now=1700000000 + ts))
-        for ts in range(3):
-            events.append(mod.build_interaction_event('LOW', ['rough_day'], 'friend', False, True, now=1700000100 + ts))
-        suggestions = mod.suggest_keyword_adjustments(events, min_observations=3)
-        by_keyword = {s['keyword']: s for s in suggestions}
-        self.assertEqual(by_keyword['want_to_die']['suggested_adjustment'], 'raise_weight')
-        self.assertEqual(by_keyword['rough_day']['suggested_adjustment'], 'lower_weight')
-
-
-class TestCrisisSynthesizerRendering(unittest.TestCase):
-    def test_render_weekly_report_outputs_json(self):
-        summary = {
-            'detections_per_level': {'NONE': 0, 'LOW': 1, 'MEDIUM': 0, 'HIGH': 0, 'CRITICAL': 0},
-            'most_common_keywords': [{'keyword': 'rough_day', 'count': 1}],
-            'false_positive_estimate': 0.0,
-            'continued_after_intervention': 0,
-            'keyword_weight_suggestions': [],
-            'window_days': 7,
-            'total_events': 1,
-        }
-        rendered = mod.render_weekly_report(summary)
-        parsed = json.loads(rendered)
-        self.assertEqual(parsed['window_days'], 7)
-        self.assertEqual(parsed['most_common_keywords'][0]['keyword'], 'rough_day')
-
-
-if __name__ == '__main__':
-    unittest.main()