diff --git a/chat/services/relationship_seed.py b/chat/services/relationship_seed.py
new file mode 100644
index 0000000..bd14592
--- /dev/null
+++ b/chat/services/relationship_seed.py
@@ -0,0 +1,107 @@
+"""Parse user-supplied "have they met?" prose into per-direction seed
+content for two bots' edges (T38).
+
+Per Requirements §5.2, when two bots first co-appear in a chat, the user
+is offered a small drawer asking "Have they met before? If yes, write a
+short prose seed describing how." That prose lands here and is parsed
+into a :class:`RelationshipSeed` whose two halves populate the
+``botA -> botB`` and ``botB -> botA`` edges respectively (summary,
+initial knowledge facts, and small affinity/trust deltas around the
+default 50/50 baseline).
+
+The two directions can differ — A may know more about B than B knows
+about A, or A may trust B less than the reverse — so the schema carries
+both halves independently.
+
+Empty/whitespace-only prose short-circuits to a default
+``RelationshipSeed`` (all zeroes, empty strings); the caller treats
+that as "they haven't met" and writes no edge content. The wrapper uses
+:func:`chat.llm.classify.classify` with ``default=RelationshipSeed()``
+so a flapping classifier degrades to the same no-op rather than
+blocking the chat-creation flow (§3.3 graceful-degradation rule).
+
+T42 (the inter-bot relationship drawer) calls this from the route layer.
+"""
+
+from __future__ import annotations
+
+from pydantic import BaseModel, Field
+
+from chat.llm.classify import classify
+from chat.llm.client import LLMClient
+
+
+class RelationshipSeed(BaseModel):
+    """Structured per-direction seed for two bots' edges.
+
+    Defaults are a deliberate no-op: empty summaries, empty knowledge
+    lists, zero deltas. Both the empty-prose short-circuit and the
+    classifier-failure fallback return this default so the caller can
+    treat them identically.
+    """
+
+    a_to_b_summary: str = ""
+    a_to_b_knowledge_facts: list[str] = Field(default_factory=list)
+    a_to_b_affinity_delta: int = 0  # signed, -10..+10 typical
+    a_to_b_trust_delta: int = 0
+    b_to_a_summary: str = ""
+    b_to_a_knowledge_facts: list[str] = Field(default_factory=list)
+    b_to_a_affinity_delta: int = 0
+    b_to_a_trust_delta: int = 0
+
+
+_SYSTEM = (
+    "You parse a short prose seed describing how two characters know each "
+    "other into structured per-direction edge content. For each direction "
+    "(A -> B, B -> A) extract: summary (one sentence from that POV), "
+    "knowledge_facts (list of factual claims that direction can carry "
+    "into future scenes), affinity_delta (-10..+10 — small adjustments to "
+    "the default 50/50 baseline), trust_delta (-10..+10). Default deltas "
+    "to 0 when prose is neutral. The two directions can differ — A may "
+    "trust B more than B trusts A. Output strict JSON matching the schema."
+)
+
+
+async def seed_inter_bot_edges(
+    client: LLMClient,
+    *,
+    classifier_model: str,
+    bot_a_id: str,
+    bot_a_name: str,
+    bot_b_id: str,
+    bot_b_name: str,
+    relationship_prose: str,
+    timeout_s: float = 30.0,
+) -> RelationshipSeed:
+    """Parse user-supplied prose into structured edge content for both
+    directed pairs.
+
+    Empty/whitespace prose short-circuits to an empty
+    :class:`RelationshipSeed` (the caller treats this as "they haven't
+    met" and writes no edge content). Classifier failure also returns
+    the default — see module docstring for the rationale.
+
+    The ``bot_a_id`` / ``bot_b_id`` arguments are accepted for symmetry
+    with the caller (T42's drawer route uses them when emitting
+    ``edge_update`` events); they're embedded in the prompt alongside
+    the names so the classifier can disambiguate when names collide.
+    """
+    if not relationship_prose or not relationship_prose.strip():
+        return RelationshipSeed()
+    user = (
+        f"Bot A: {bot_a_name} (id={bot_a_id})\n"
+        f"Bot B: {bot_b_name} (id={bot_b_id})\n\n"
+        f"Prose seed:\n{relationship_prose.strip()}"
+    )
+    return await classify(
+        client,
+        model=classifier_model,
+        system=_SYSTEM,
+        user=user,
+        schema=RelationshipSeed,
+        default=RelationshipSeed(),
+        timeout_s=timeout_s,
+    )
+
+
+__all__ = ["RelationshipSeed", "seed_inter_bot_edges"]
diff --git a/tests/test_relationship_seed.py b/tests/test_relationship_seed.py
new file mode 100644
index 0000000..b2b7312
--- /dev/null
+++ b/tests/test_relationship_seed.py
@@ -0,0 +1,109 @@
+"""Tests for the relationship-seed service (T38).
+
+Per Requirements §5.2, when two bots first co-appear in a chat, the user
+is prompted with "Have they met before? If yes, write a short prose
+seed." The prose is parsed via classifier into structured directed-edge
+content for the ``botA -> botB`` and ``botB -> botA`` edges.
+
+These tests cover:
+
+* The happy path: a canned classifier response parses cleanly into a
+  populated :class:`RelationshipSeed` with both directions filled.
+* Empty prose short-circuits before any classifier call (mock has no
+  canned responses; an accidental call would raise ``IndexError``).
+* Whitespace-only prose has the same short-circuit behavior.
+"""
+
+from __future__ import annotations
+
+import json
+
+import pytest
+
+from chat.llm.mock import MockLLMClient
+from chat.services.relationship_seed import (
+    RelationshipSeed,
+    seed_inter_bot_edges,
+)
+
+
+@pytest.mark.asyncio
+async def test_seed_parses_canned_prose():
+    canned = json.dumps(
+        {
+            "a_to_b_summary": "old college friend who now distrusts him slightly",
+            "a_to_b_knowledge_facts": [
+                "studied physics together",
+                "lost touch after a falling out",
+            ],
+            "a_to_b_affinity_delta": 2,
+            "a_to_b_trust_delta": -1,
+            "b_to_a_summary": "former roommate; warm memories, mild resentment",
+            "b_to_a_knowledge_facts": ["lived together junior year"],
+            "b_to_a_affinity_delta": 3,
+            "b_to_a_trust_delta": 0,
+        }
+    )
+    mock = MockLLMClient(canned=[canned])
+    result = await seed_inter_bot_edges(
+        mock,
+        classifier_model="x",
+        bot_a_id="bot_a",
+        bot_a_name="Alice",
+        bot_b_id="bot_b",
+        bot_b_name="Bob",
+        relationship_prose=(
+            "Alice and Bob met in college. They studied physics together and "
+            "lived as roommates junior year, but drifted apart after a fight."
+        ),
+    )
+    assert isinstance(result, RelationshipSeed)
+    assert (
+        result.a_to_b_summary
+        == "old college friend who now distrusts him slightly"
+    )
+    assert result.a_to_b_knowledge_facts == [
+        "studied physics together",
+        "lost touch after a falling out",
+    ]
+    assert result.a_to_b_affinity_delta == 2
+    assert result.a_to_b_trust_delta == -1
+    assert (
+        result.b_to_a_summary
+        == "former roommate; warm memories, mild resentment"
+    )
+    assert result.b_to_a_knowledge_facts == ["lived together junior year"]
+    assert result.b_to_a_affinity_delta == 3
+    assert result.b_to_a_trust_delta == 0
+
+
+@pytest.mark.asyncio
+async def test_seed_empty_prose_returns_empty():
+    """Empty prose short-circuits — classifier must not be called."""
+    mock = MockLLMClient(canned=[])
+    result = await seed_inter_bot_edges(
+        mock,
+        classifier_model="x",
+        bot_a_id="bot_a",
+        bot_a_name="Alice",
+        bot_b_id="bot_b",
+        bot_b_name="Bob",
+        relationship_prose="",
+    )
+    assert result == RelationshipSeed()
+
+
+@pytest.mark.asyncio
+async def test_seed_whitespace_only_prose_returns_empty():
+    """Whitespace-only prose is treated the same as empty."""
+    mock = MockLLMClient(canned=[])
+    result = await seed_inter_bot_edges(
+        mock,
+        classifier_model="x",
+        bot_a_id="bot_a",
+        bot_a_name="Alice",
+        bot_b_id="bot_b",
+        bot_b_name="Bob",
+        relationship_prose="   \n  ",
+    )
+    assert result == RelationshipSeed()