fix

.gitignore

@@ -312,3 +312,4 @@ cython_debug/
 # postgres data
 data_postgres/
 
+secrets/

athena/app/services/episodic_memory_storage_service.py

@@ -67,26 +67,26 @@ async def _upsert_one(self, session: AsyncSession, unit: EpisodicMemoryUnit) ->
         # Check existing
         existing = await session.scalar(
             select(EpisodicMemoryUnitDB).where(
-                col(EpisodicMemoryUnitDB.memory_id) == unit.memory_id
+                col(EpisodicMemoryUnitDB.episodic_memory_id) == unit.memory_id
             )
         )
 
         if existing is None:
-            row = EpisodicMemoryUnitDB.from_memory_unit(unit)
+            row = EpisodicMemoryUnitDB.from_episodic_memory_unit(unit)
             row.task_embedding = task_vec
             row.state_embedding = state_vec
             row.task_state_embedding = task_state_vec
             session.add(row)
         else:
             # Update fields
-            fresh = EpisodicMemoryUnitDB.from_memory_unit(unit)
+            fresh = EpisodicMemoryUnitDB.from_episodic_memory_unit(unit)
             for attr in (
-                "memory_created_at",
-                "memory_updated_at",
-                "memory_invalid_at",
-                "memory_source_name",
-                "memory_run_id",
-                "memory_metadata",
+                "episodic_memory_created_at",
+                "episodic_memory_updated_at",
+                "episodic_memory_invalid_at",
+                "episodic_memory_source_name",
+                "episodic_memory_run_id",
+                "episodic_memory_metadata",
                 "task_issue_title",
                 "task_issue_body",
                 "task_issue_comments",
@@ -138,14 +138,16 @@ async def search_by_text(
                 .limit(limit)
             )
             rows: List[EpisodicMemoryUnitDB] = list(res.scalars())
-        return [r.to_memory_unit() for r in rows]
+        return [r.to_episodic_memory_unit() for r in rows]
 
     async def get_by_memory_id(self, memory_id: str) -> Optional[EpisodicMemoryUnit]:
         async with self._sessionmaker() as session:
             row = await session.scalar(
-                select(EpisodicMemoryUnitDB).where(col(EpisodicMemoryUnitDB.memory_id) == memory_id)
+                select(EpisodicMemoryUnitDB).where(
+                    col(EpisodicMemoryUnitDB.episodic_memory_id) == memory_id
+                )
             )
-            return None if row is None else row.to_memory_unit()
+            return None if row is None else row.to_episodic_memory_unit()
 
     async def list_all(self, limit: Optional[int] = None) -> List[EpisodicMemoryUnit]:
         async with self._sessionmaker() as session:
@@ -154,12 +156,12 @@ async def list_all(self, limit: Optional[int] = None) -> List[EpisodicMemoryUnit
                 stmt = stmt.limit(limit)
             res = await session.execute(stmt)
             rows: List[EpisodicMemoryUnitDB] = list(res.scalars())
-        return [r.to_memory_unit() for r in rows]
+        return [r.to_episodic_memory_unit() for r in rows]
 
     async def clear_all(self) -> None:
         async with self._sessionmaker() as session:
             # Delete all rows from table
-            await session.execute("DELETE FROM memory_units")
+            await session.execute("DELETE FROM episodic_memories")
             await session.commit()
 
     @staticmethod

athena/app/services/llm_service.py

@@ -51,7 +51,6 @@ def get_model(
     elif model_name.startswith("vertex:"):
         # example: model_name="vertex:gemini-2.5-pro"
         vertex_model = model_name.split("vertex:", 1)[1]
-        print(google_application_credentials)
         if isinstance(google_application_credentials, dict):
             creds = Credentials.from_service_account_info(
                 google_application_credentials,

docker-compose.yml

@@ -43,6 +43,7 @@ services:
       - ATHENA_GEMINI_API_KEY=${ATHENA_GEMINI_API_KEY}
       - ATHENA_OPENAI_FORMAT_API_KEY=${ATHENA_OPENAI_FORMAT_API_KEY}
       - ATHENA_OPENAI_FORMAT_BASE_URL=${ATHENA_OPENAI_FORMAT_BASE_URL}
+      - ATHENA_GOOGLE_APPLICATION_CREDENTIALS=${ATHENA_GOOGLE_APPLICATION_CREDENTIALS}
 
       # Model settings
       - ATHENA_MODEL_TEMPERATURE=${ATHENA_MODEL_TEMPERATURE}
@@ -57,6 +58,7 @@ services:
     volumes:
       - .:/app
       - /var/run/docker.sock:/var/run/docker.sock
+      - ./secrets/prometheus-code-agent.json:/secrets/prometheus-code-agent.json:ro
     depends_on:
       postgres:
         condition: service_healthy

example.env

@@ -6,7 +6,7 @@ ATHENA_ENVIRONMENT=local
 ATHENA_BACKEND_CORS_ORIGINS=["*"]
 
 # LLM model settings
-ATHENA_MODEL_NAME=gpt-4o
+ATHENA_MODEL_NAME=vertex:gemini-2.5-flash
 
 # API keys for various LLM providers
 ATHENA_ANTHROPIC_API_KEY=anthropic_api_key