hindsight-api 0.2.1__py3-none-any.whl → 0.4.0__py3-none-any.whl

hindsight_api/extensions/tenant.py

@@ -28,6 +28,18 @@ class TenantContext:
     schema_name: str
 
 
+@dataclass
+class Tenant:
+    """
+    Represents a tenant for worker discovery.
+
+    Used by list_tenants() to return tenant information including
+    the PostgreSQL schema name for database operations.
+    """
+
+    schema: str
+
+
 class TenantExtension(Extension, ABC):
     """
     Extension for multi-tenancy and API key authentication.
@@ -61,3 +73,17 @@ class TenantExtension(Extension, ABC):
             AuthenticationError: If authentication fails.
         """
         ...
+
+    @abstractmethod
+    async def list_tenants(self) -> list[Tenant]:
+        """
+        List all tenants that should be processed by workers.
+
+        This method is used by the worker to discover all tenants that need
+        task polling. Workers will poll for pending tasks in each tenant's schema.
+
+        Returns:
+            List of Tenant objects containing schema information.
+            For single-tenant setups, return [Tenant(schema="public")].
+        """
+        ...

hindsight_api/main.py

@@ -23,7 +23,7 @@ import uvicorn
 from . import MemoryEngine
 from .api import create_app
 from .banner import print_banner
-from .config import HindsightConfig, get_config
+from .config import DEFAULT_WORKERS, ENV_WORKERS, HindsightConfig, get_config
 from .daemon import (
     DEFAULT_DAEMON_PORT,
     DEFAULT_IDLE_TIMEOUT,
@@ -95,7 +95,12 @@ def main():
 
     # Development options
     parser.add_argument("--reload", action="store_true", help="Enable auto-reload on code changes (development only)")
-    parser.add_argument("--workers", type=int, default=1, help="Number of worker processes (default: 1)")
+    parser.add_argument(
+        "--workers",
+        type=int,
+        default=int(os.getenv(ENV_WORKERS, str(DEFAULT_WORKERS))),
+        help=f"Number of worker processes (env: {ENV_WORKERS}, default: {DEFAULT_WORKERS})",
+    )
 
     # Access log options
     parser.add_argument("--access-log", action="store_true", help="Enable access log")
@@ -171,21 +176,62 @@ def main():
             llm_base_url=config.llm_base_url,
             llm_max_concurrent=config.llm_max_concurrent,
             llm_timeout=config.llm_timeout,
+            retain_llm_provider=config.retain_llm_provider,
+            retain_llm_api_key=config.retain_llm_api_key,
+            retain_llm_model=config.retain_llm_model,
+            retain_llm_base_url=config.retain_llm_base_url,
+            reflect_llm_provider=config.reflect_llm_provider,
+            reflect_llm_api_key=config.reflect_llm_api_key,
+            reflect_llm_model=config.reflect_llm_model,
+            reflect_llm_base_url=config.reflect_llm_base_url,
+            consolidation_llm_provider=config.consolidation_llm_provider,
+            consolidation_llm_api_key=config.consolidation_llm_api_key,
+            consolidation_llm_model=config.consolidation_llm_model,
+            consolidation_llm_base_url=config.consolidation_llm_base_url,
             embeddings_provider=config.embeddings_provider,
             embeddings_local_model=config.embeddings_local_model,
             embeddings_tei_url=config.embeddings_tei_url,
+            embeddings_openai_base_url=config.embeddings_openai_base_url,
+            embeddings_cohere_base_url=config.embeddings_cohere_base_url,
             reranker_provider=config.reranker_provider,
             reranker_local_model=config.reranker_local_model,
             reranker_tei_url=config.reranker_tei_url,
+            reranker_tei_batch_size=config.reranker_tei_batch_size,
+            reranker_tei_max_concurrent=config.reranker_tei_max_concurrent,
+            reranker_max_candidates=config.reranker_max_candidates,
+            reranker_cohere_base_url=config.reranker_cohere_base_url,
             host=args.host,
             port=args.port,
             log_level=args.log_level,
+            log_format=config.log_format,
             mcp_enabled=config.mcp_enabled,
             graph_retriever=config.graph_retriever,
-            observation_min_facts=config.observation_min_facts,
-            observation_top_entities=config.observation_top_entities,
+            mpfp_top_k_neighbors=config.mpfp_top_k_neighbors,
+            recall_max_concurrent=config.recall_max_concurrent,
+            recall_connection_budget=config.recall_connection_budget,
+            retain_max_completion_tokens=config.retain_max_completion_tokens,
+            retain_chunk_size=config.retain_chunk_size,
+            retain_extract_causal_links=config.retain_extract_causal_links,
+            retain_extraction_mode=config.retain_extraction_mode,
+            retain_custom_instructions=config.retain_custom_instructions,
+            retain_observations_async=config.retain_observations_async,
+            enable_observations=config.enable_observations,
+            consolidation_batch_size=config.consolidation_batch_size,
             skip_llm_verification=config.skip_llm_verification,
             lazy_reranker=config.lazy_reranker,
+            run_migrations_on_startup=config.run_migrations_on_startup,
+            db_pool_min_size=config.db_pool_min_size,
+            db_pool_max_size=config.db_pool_max_size,
+            db_command_timeout=config.db_command_timeout,
+            db_acquire_timeout=config.db_acquire_timeout,
+            worker_enabled=config.worker_enabled,
+            worker_id=config.worker_id,
+            worker_poll_interval_ms=config.worker_poll_interval_ms,
+            worker_max_retries=config.worker_max_retries,
+            worker_batch_size=config.worker_batch_size,
+            worker_http_port=config.worker_http_port,
+            reflect_max_iterations=config.reflect_max_iterations,
+            mental_model_refresh_concurrency=config.mental_model_refresh_concurrency,
         )
     config.configure_logging()
     if not args.daemon:
@@ -211,7 +257,11 @@ def main():
         logging.info(f"Loaded tenant extension: {tenant_extension.__class__.__name__}")
 
     # Create MemoryEngine (reads configuration from environment)
-    _memory = MemoryEngine(operation_validator=operation_validator, tenant_extension=tenant_extension)
+    _memory = MemoryEngine(
+        operation_validator=operation_validator,
+        tenant_extension=tenant_extension,
+        run_migrations=config.run_migrations_on_startup,
+    )
 
     # Set extension context on tenant extension (needed for schema provisioning)
     if tenant_extension:
@@ -238,14 +288,27 @@ def main():
         app = idle_middleware
 
     # Prepare uvicorn config
+    # When using workers or reload, we must use import string so each worker can import the app
+    use_import_string = args.workers > 1 or args.reload
+    # Check for uvloop availability
+    try:
+        import uvloop  # noqa: F401
+
+        loop_impl = "uvloop"
+        print("uvloop available, will use for event loop")
+    except ImportError:
+        loop_impl = "asyncio"
+        print("uvloop not installed, using default asyncio event loop")
+
     uvicorn_config = {
-        "app": app,
+        "app": "hindsight_api.server:app" if use_import_string else app,
         "host": args.host,
         "port": args.port,
         "log_level": args.log_level,
         "access_log": args.access_log,
         "proxy_headers": args.proxy_headers,
         "ws": "wsproto",  # Use wsproto instead of websockets to avoid deprecation warnings
+        "loop": loop_impl,  # Explicitly set event loop implementation
     }
 
     # Add optional parameters if provided

hindsight_api/mcp_local.py

@@ -44,7 +44,6 @@ import os
 import sys
 
 from mcp.server.fastmcp import FastMCP
-from mcp.types import Icon
 
 from hindsight_api.config import (
     DEFAULT_MCP_LOCAL_BANK_ID,
@@ -53,6 +52,7 @@ from hindsight_api.config import (
     ENV_MCP_INSTRUCTIONS,
     ENV_MCP_LOCAL_BANK_ID,
 )
+from hindsight_api.mcp_tools import MCPToolsConfig, register_mcp_tools
 
 # Configure logging - default to warning to avoid polluting stderr during MCP init
 # MCP clients interpret stderr output as errors, so we suppress INFO logs by default
@@ -85,9 +85,6 @@ def create_local_mcp_server(bank_id: str, memory=None) -> FastMCP:
     """
     # Import here to avoid slow startup if just checking --help
     from hindsight_api import MemoryEngine
-    from hindsight_api.engine.memory_engine import Budget
-    from hindsight_api.engine.response_models import VALID_RECALL_FACT_TYPES
-    from hindsight_api.models import RequestContext
 
     # Create memory engine with pg0 embedded database if not provided
     if memory is None:
@@ -105,55 +102,17 @@ def create_local_mcp_server(bank_id: str, memory=None) -> FastMCP:
 
     mcp = FastMCP("hindsight")
 
-    @mcp.tool(description=retain_description)
-    async def retain(content: str, context: str = "general") -> dict:
-        """
-        Args:
-            content: The fact/memory to store (be specific and include relevant details)
-            context: Category for the memory (e.g., 'preferences', 'work', 'hobbies', 'family'). Default: 'general'
-        """
-        import asyncio
-
-        async def _retain():
-            try:
-                await memory.retain_batch_async(
-                    bank_id=bank_id,
-                    contents=[{"content": content, "context": context}],
-                    request_context=RequestContext(),
-                )
-            except Exception as e:
-                logger.error(f"Error storing memory: {e}", exc_info=True)
-
-        # Fire and forget - don't block on memory storage
-        asyncio.create_task(_retain())
-        return {"status": "accepted", "message": "Memory storage initiated"}
-
-    @mcp.tool(description=recall_description)
-    async def recall(query: str, max_tokens: int = 4096, budget: str = "low") -> dict:
-        """
-        Args:
-            query: Natural language search query (e.g., "user's food preferences", "what projects is user working on")
-            max_tokens: Maximum tokens to return in results (default: 4096)
-            budget: Search budget level - "low", "mid", or "high" (default: "low")
-        """
-        try:
-            # Map string budget to enum
-            budget_map = {"low": Budget.LOW, "mid": Budget.MID, "high": Budget.HIGH}
-            budget_enum = budget_map.get(budget.lower(), Budget.LOW)
-
-            search_result = await memory.recall_async(
-                bank_id=bank_id,
-                query=query,
-                fact_type=list(VALID_RECALL_FACT_TYPES),
-                budget=budget_enum,
-                max_tokens=max_tokens,
-                request_context=RequestContext(),
-            )
-
-            return search_result.model_dump()
-        except Exception as e:
-            logger.error(f"Error searching: {e}", exc_info=True)
-            return {"error": str(e), "results": []}
+    # Configure and register tools using shared module
+    config = MCPToolsConfig(
+        bank_id_resolver=lambda: bank_id,
+        include_bank_id_param=False,  # Local MCP uses fixed bank_id
+        tools={"retain", "recall"},  # Local MCP only has retain and recall
+        retain_description=retain_description,
+        recall_description=recall_description,
+        retain_fire_and_forget=True,  # Local MCP uses fire-and-forget pattern
+    )
+
+    register_mcp_tools(mcp, memory, config)
 
     return mcp