feat: synthesized-memories service for jump skips (T54)

chat/services/synthesized_memories.py

@@ -0,0 +1,74 @@
+"""Synthesized-memories service (T54).
+
+When the user jump-skips with 'anything notable happen?' prose, parse
+that prose into 1-N synthesized memories per present bot. Each memory
+carries source="synthesized" and reliability=0.7 (lower than direct).
+Caller (T62 skip flow) writes the memories via record_turn_memory_for_present.
+"""
+
+from __future__ import annotations
+
+from pydantic import BaseModel, Field
+
+from chat.llm.classify import classify
+from chat.llm.client import LLMClient
+
+
+class SynthesizedMemory(BaseModel):
+    text: str
+    significance: int = 1   # 0..3, default 1
+    affinity_delta: int = 0
+    trust_delta: int = 0
+
+
+class SynthesizedDigest(BaseModel):
+    memories: list[SynthesizedMemory] = Field(default_factory=list)
+
+
+_SYSTEM = (
+    "You parse a short user-supplied prose describing 'anything notable' "
+    "that happened during a time skip into 1-N synthesized memories from "
+    "a single bot's POV. Each memory has: text (one factual sentence "
+    "from that bot's perspective), significance (0-3, default 1; only "
+    "use 2 or 3 for genuinely scene-level or relationship-altering "
+    "events), affinity_delta and trust_delta (-10..+10, default 0; "
+    "use small adjustments only when prose explicitly describes a shift). "
+    "Empty/whitespace prose returns an empty memories list. Output "
+    "strict JSON matching the schema."
+)
+
+
+async def synthesize_memories(
+    client: LLMClient,
+    *,
+    classifier_model: str,
+    prose: str,
+    bot_name: str,
+    bot_persona: str,
+    you_name: str,
+    timeout_s: float = 30.0,
+) -> SynthesizedDigest:
+    """Parse 'anything notable' prose into structured memories from a
+    single bot's POV. Empty/whitespace prose short-circuits to an
+    empty digest (no LLM call)."""
+    if not prose or not prose.strip():
+        return SynthesizedDigest()
+
+    user = (
+        f"Bot: {bot_name}\n"
+        f"Persona: {bot_persona}\n"
+        f"Other party: {you_name}\n\n"
+        f"Prose:\n{prose.strip()}"
+    )
+    return await classify(
+        client,
+        model=classifier_model,
+        system=_SYSTEM,
+        user=user,
+        schema=SynthesizedDigest,
+        default=SynthesizedDigest(),
+        timeout_s=timeout_s,
+    )
+
+
+__all__ = ["SynthesizedMemory", "SynthesizedDigest", "synthesize_memories"]

tests/test_synthesized_memories.py

@@ -0,0 +1,98 @@
+"""Tests for the synthesized-memories service (T54).
+
+When the user jump-skips ("a week later") they are prompted "anything
+notable happen?" If they answer with prose, this service parses it into
+1-N synthesized memories per present bot. Each memory carries
+``source="synthesized"`` and ``reliability=0.7`` (the caller — T62 skip
+flow — applies those tags when persisting; this service just produces
+the structured digest).
+
+These tests cover:
+
+* The happy path: a canned classifier response parses cleanly into a
+  populated :class:`SynthesizedDigest` with one memory.
+* Empty prose short-circuits before any classifier call — the mock has
+  no canned responses, so an accidental call would raise
+  ``IndexError``.
+* Classifier failure (3 bad responses, exhausting :func:`classify`'s
+  retry budget) falls back to an empty default digest.
+"""
+
+from __future__ import annotations
+
+import json
+
+import pytest
+
+from chat.llm.mock import MockLLMClient
+from chat.services.synthesized_memories import (
+    SynthesizedDigest,
+    SynthesizedMemory,
+    synthesize_memories,
+)
+
+
+@pytest.mark.asyncio
+async def test_synthesize_parses_canned_prose():
+    canned = json.dumps(
+        {
+            "memories": [
+                {
+                    "text": "Maya started a new pottery class.",
+                    "significance": 1,
+                    "affinity_delta": 0,
+                    "trust_delta": 0,
+                }
+            ]
+        }
+    )
+    mock = MockLLMClient(canned=[canned])
+    result = await synthesize_memories(
+        mock,
+        classifier_model="x",
+        prose="we saw each other at her pottery class once",
+        bot_name="Maya",
+        bot_persona="warm potter, mid-30s",
+        you_name="Sam",
+    )
+    assert isinstance(result, SynthesizedDigest)
+    assert len(result.memories) == 1
+    mem = result.memories[0]
+    assert isinstance(mem, SynthesizedMemory)
+    assert mem.text == "Maya started a new pottery class."
+    assert mem.significance == 1
+    assert mem.affinity_delta == 0
+    assert mem.trust_delta == 0
+
+
+@pytest.mark.asyncio
+async def test_empty_prose_returns_empty_digest():
+    """Empty prose short-circuits — the classifier must not be called."""
+    mock = MockLLMClient(canned=[])
+    result = await synthesize_memories(
+        mock,
+        classifier_model="x",
+        prose="",
+        bot_name="Maya",
+        bot_persona="warm potter, mid-30s",
+        you_name="Sam",
+    )
+    assert result == SynthesizedDigest()
+    assert result.memories == []
+
+
+@pytest.mark.asyncio
+async def test_classifier_failure_returns_empty_default():
+    """Three bad responses exhaust the classifier's retry budget; the
+    service then returns the empty default digest."""
+    mock = MockLLMClient(canned=["bad", "bad", "bad"])
+    result = await synthesize_memories(
+        mock,
+        classifier_model="x",
+        prose="we saw each other at her pottery class once",
+        bot_name="Maya",
+        bot_persona="warm potter, mid-30s",
+        you_name="Sam",
+    )
+    assert result == SynthesizedDigest()
+    assert result.memories == []