hindsight-api 0.4.7__py3-none-any.whl → 0.4.8__py3-none-any.whl

hindsight_api/engine/memory_engine.py

@@ -303,8 +303,10 @@ class MemoryEngine(MemoryEngineInterface):
         db_url = db_url or config.database_url
         memory_llm_provider = memory_llm_provider or config.llm_provider
         memory_llm_api_key = memory_llm_api_key or config.llm_api_key
-        # Ollama and mock don't require an API key
-        if not memory_llm_api_key and memory_llm_provider not in ("ollama", "mock"):
+        # Ollama, openai-codex, claude-code, and mock don't require an API key
+        # openai-codex uses OAuth tokens from ~/.codex/auth.json
+        # claude-code uses OAuth tokens from macOS Keychain
+        if not memory_llm_api_key and memory_llm_provider not in ("ollama", "openai-codex", "claude-code", "mock"):
             raise ValueError("LLM API key is required. Set HINDSIGHT_API_LLM_API_KEY environment variable.")
         memory_llm_model = memory_llm_model or config.llm_model
         memory_llm_base_url = memory_llm_base_url or config.get_llm_base_url() or None
@@ -457,7 +459,11 @@ class MemoryEngine(MemoryEngineInterface):
         # Store operation validator extension (optional)
         self._operation_validator = operation_validator
 
-        # Store tenant extension (optional)
+        # Store tenant extension (always set, use default if none provided)
+        if tenant_extension is None:
+            from ..extensions.builtin.tenant import DefaultTenantExtension
+
+            tenant_extension = DefaultTenantExtension(config={})
         self._tenant_extension = tenant_extension
 
     async def _validate_operation(self, validation_coro) -> None:
@@ -495,22 +501,18 @@ class MemoryEngine(MemoryEngineInterface):
         Raises:
             AuthenticationError: If authentication fails or request_context is missing when required.
         """
-        if self._tenant_extension is None:
-            _current_schema.set("public")
-            return "public"
-
         from hindsight_api.extensions import AuthenticationError
 
         if request_context is None:
-            raise AuthenticationError("RequestContext is required when tenant extension is configured")
+            raise AuthenticationError("RequestContext is required")
 
         # For internal/background operations (e.g., worker tasks), skip extension authentication.
         # The task was already authenticated at submission time, and execute_task sets _current_schema
-        # from the task's _schema field. For public schema tasks, _current_schema keeps its default "public".
+        # from the task's _schema field.
         if request_context.internal:
             return _current_schema.get()
 
-        # Let AuthenticationError propagate - HTTP layer will convert to 401
+        # Authenticate through tenant extension (always set, may be default no-auth extension)
         tenant_context = await self._tenant_extension.authenticate(request_context)
 
         _current_schema.set(tenant_context.schema_name)
@@ -536,10 +538,15 @@ class MemoryEngine(MemoryEngineInterface):
             f"[BATCH_RETAIN_TASK] Starting background batch retain for bank_id={bank_id}, {len(contents)} items"
         )
 
-        # Use internal request context for background tasks (skips tenant auth when schema is pre-set)
+        # Restore tenant_id/api_key_id from task payload so downstream operations
+        # (e.g., consolidation and mental model refreshes) can attribute usage.
         from hindsight_api.models import RequestContext
 
-        internal_context = RequestContext(internal=True)
+        internal_context = RequestContext(
+            internal=True,
+            tenant_id=task_dict.get("_tenant_id"),
+            api_key_id=task_dict.get("_api_key_id"),
+        )
         await self.retain_batch_async(bank_id=bank_id, contents=contents, request_context=internal_context)
 
         logger.info(f"[BATCH_RETAIN_TASK] Completed background batch retain for bank_id={bank_id}")
@@ -565,7 +572,13 @@ class MemoryEngine(MemoryEngineInterface):
 
         from .consolidation import run_consolidation_job
 
-        internal_context = RequestContext(internal=True)
+        # Restore tenant_id/api_key_id from task payload so downstream operations
+        # (e.g., mental model refreshes) can attribute usage to the correct org.
+        internal_context = RequestContext(
+            internal=True,
+            tenant_id=task_dict.get("_tenant_id"),
+            api_key_id=task_dict.get("_api_key_id"),
+        )
         result = await run_consolidation_job(
             memory_engine=self,
             bank_id=bank_id,
@@ -926,30 +939,34 @@ class MemoryEngine(MemoryEngineInterface):
 
             if not self.db_url:
                 raise ValueError("Database URL is required for migrations")
-            logger.info("Running database migrations...")
-            # Use configured database schema for migrations (defaults to "public")
-            run_migrations(self.db_url, schema=get_config().database_schema)
 
-            # Migrate all existing tenant schemas (if multi-tenant)
-            if self._tenant_extension is not None:
-                try:
-                    tenants = await self._tenant_extension.list_tenants()
-                    if tenants:
-                        logger.info(f"Running migrations on {len(tenants)} tenant schemas...")
-                        for tenant in tenants:
-                            schema = tenant.schema
-                            if schema and schema != "public":
-                                try:
-                                    run_migrations(self.db_url, schema=schema)
-                                except Exception as e:
-                                    logger.warning(f"Failed to migrate tenant schema {schema}: {e}")
-                        logger.info("Tenant schema migrations completed")
-                except Exception as e:
-                    logger.warning(f"Failed to run tenant schema migrations: {e}")
-
-            # Ensure embedding column dimension matches the model's dimension
-            # This is done after migrations and after embeddings.initialize()
-            ensure_embedding_dimension(self.db_url, self.embeddings.dimension, schema=get_config().database_schema)
+            # Migrate all schemas from the tenant extension
+            # The tenant extension is the single source of truth for which schemas exist
+            logger.info("Running database migrations...")
+            try:
+                tenants = await self._tenant_extension.list_tenants()
+                if tenants:
+                    logger.info(f"Running migrations on {len(tenants)} schema(s)...")
+                    for tenant in tenants:
+                        schema = tenant.schema
+                        if schema:
+                            try:
+                                run_migrations(self.db_url, schema=schema)
+                            except Exception as e:
+                                logger.warning(f"Failed to migrate schema {schema}: {e}")
+                    logger.info("Schema migrations completed")
+
+                    # Ensure embedding column dimension matches the model's dimension
+                    # This is done after migrations and after embeddings.initialize()
+                    for tenant in tenants:
+                        schema = tenant.schema
+                        if schema:
+                            try:
+                                ensure_embedding_dimension(self.db_url, self.embeddings.dimension, schema=schema)
+                            except Exception as e:
+                                logger.warning(f"Failed to ensure embedding dimension for schema {schema}: {e}")
+            except Exception as e:
+                logger.warning(f"Failed to run schema migrations: {e}")
 
         logger.info(f"Connecting to PostgreSQL at {self.db_url}")
 
@@ -5458,6 +5475,13 @@ class MemoryEngine(MemoryEngineInterface):
         task_payload: dict[str, Any] = {"contents": contents}
         if document_tags:
             task_payload["document_tags"] = document_tags
+        # Pass tenant_id and api_key_id through task payload so the worker
+        # can propagate request context to downstream operations (e.g.,
+        # consolidation and mental model refreshes triggered after retain).
+        if request_context.tenant_id:
+            task_payload["_tenant_id"] = request_context.tenant_id
+        if request_context.api_key_id:
+            task_payload["_api_key_id"] = request_context.api_key_id
 
         result = await self._submit_async_operation(
             bank_id=bank_id,
@@ -5490,11 +5514,21 @@ class MemoryEngine(MemoryEngineInterface):
             Dict with operation_id
         """
         await self._authenticate_tenant(request_context)
+
+        # Pass tenant_id and api_key_id through task payload so the worker
+        # can provide request context to extension hooks (e.g., usage metering
+        # for mental model refreshes triggered by consolidation).
+        task_payload: dict[str, Any] = {}
+        if request_context.tenant_id:
+            task_payload["_tenant_id"] = request_context.tenant_id
+        if request_context.api_key_id:
+            task_payload["_api_key_id"] = request_context.api_key_id
+
         return await self._submit_async_operation(
             bank_id=bank_id,
             operation_type="consolidation",
             task_type="consolidation",
-            task_payload={},
+            task_payload=task_payload,
             dedupe_by_bank=True,
         )
 

hindsight_api/engine/providers/__init__.py

@@ -0,0 +1,14 @@
+"""
+LLM provider implementations.
+
+This package contains concrete implementations of the LLMInterface for various providers.
+"""
+
+from .anthropic_llm import AnthropicLLM
+from .claude_code_llm import ClaudeCodeLLM
+from .codex_llm import CodexLLM
+from .gemini_llm import GeminiLLM
+from .mock_llm import MockLLM
+from .openai_compatible_llm import OpenAICompatibleLLM
+
+__all__ = ["AnthropicLLM", "ClaudeCodeLLM", "CodexLLM", "GeminiLLM", "MockLLM", "OpenAICompatibleLLM"]

hindsight_api/engine/providers/anthropic_llm.py

@@ -0,0 +1,434 @@
+"""
+Anthropic LLM provider using the Anthropic Python SDK.
+
+This provider enables using Claude models from Anthropic with support for:
+- Structured JSON output
+- Tool/function calling with proper format conversion
+- Extended thinking mode
+- Retry logic with exponential backoff
+"""
+
+import asyncio
+import json
+import logging
+import time
+from typing import Any
+
+from hindsight_api.engine.llm_interface import LLMInterface, OutputTooLongError
+from hindsight_api.engine.response_models import LLMToolCall, LLMToolCallResult, TokenUsage
+from hindsight_api.metrics import get_metrics_collector
+
+logger = logging.getLogger(__name__)
+
+
+class AnthropicLLM(LLMInterface):
+    """
+    LLM provider using Anthropic's Claude models.
+
+    Supports structured output, tool calling, and extended thinking mode.
+    Handles format conversion between OpenAI-style messages and Anthropic's format.
+    """
+
+    def __init__(
+        self,
+        provider: str,
+        api_key: str,
+        base_url: str,
+        model: str,
+        reasoning_effort: str = "low",
+        timeout: float = 300.0,
+        **kwargs: Any,
+    ):
+        """
+        Initialize Anthropic LLM provider.
+
+        Args:
+            provider: Provider name (should be "anthropic").
+            api_key: Anthropic API key.
+            base_url: Base URL for the API (optional, uses Anthropic default if empty).
+            model: Model name (e.g., "claude-sonnet-4-20250514").
+            reasoning_effort: Reasoning effort level (not used by Anthropic).
+            timeout: Request timeout in seconds.
+            **kwargs: Additional provider-specific parameters.
+        """
+        super().__init__(provider, api_key, base_url, model, reasoning_effort, **kwargs)
+
+        if not self.api_key:
+            raise ValueError("API key is required for Anthropic provider")
+
+        # Import and initialize Anthropic client
+        try:
+            from anthropic import AsyncAnthropic
+
+            client_kwargs: dict[str, Any] = {"api_key": self.api_key}
+            if self.base_url:
+                client_kwargs["base_url"] = self.base_url
+            if timeout:
+                client_kwargs["timeout"] = timeout
+
+            self._client = AsyncAnthropic(**client_kwargs)
+            logger.info(f"Anthropic client initialized for model: {self.model}")
+        except ImportError as e:
+            raise RuntimeError("Anthropic SDK not installed. Run: uv add anthropic or pip install anthropic") from e
+
+    async def verify_connection(self) -> None:
+        """
+        Verify that the Anthropic provider is configured correctly by making a simple test call.
+
+        Raises:
+            RuntimeError: If the connection test fails.
+        """
+        try:
+            test_messages = [{"role": "user", "content": "test"}]
+            await self.call(
+                messages=test_messages,
+                max_completion_tokens=10,
+                temperature=0.0,
+                scope="test",
+                max_retries=0,
+            )
+            logger.info("Anthropic connection verified successfully")
+        except Exception as e:
+            logger.error(f"Anthropic connection verification failed: {e}")
+            raise RuntimeError(f"Failed to verify Anthropic connection: {e}") from e
+
+    async def call(
+        self,
+        messages: list[dict[str, str]],
+        response_format: Any | None = None,
+        max_completion_tokens: int | None = None,
+        temperature: float | None = None,
+        scope: str = "memory",
+        max_retries: int = 10,
+        initial_backoff: float = 1.0,
+        max_backoff: float = 60.0,
+        skip_validation: bool = False,
+        strict_schema: bool = False,
+        return_usage: bool = False,
+    ) -> Any:
+        """
+        Make an LLM API call with retry logic.
+
+        Args:
+            messages: List of message dicts with 'role' and 'content'.
+            response_format: Optional Pydantic model for structured output.
+            max_completion_tokens: Maximum tokens in response.
+            temperature: Sampling temperature (0.0-2.0).
+            scope: Scope identifier for tracking.
+            max_retries: Maximum retry attempts.
+            initial_backoff: Initial backoff time in seconds.
+            max_backoff: Maximum backoff time in seconds.
+            skip_validation: Return raw JSON without Pydantic validation.
+            strict_schema: Use strict JSON schema enforcement (not supported by Anthropic).
+            return_usage: If True, return tuple (result, TokenUsage) instead of just result.
+
+        Returns:
+            If return_usage=False: Parsed response if response_format is provided, otherwise text content.
+            If return_usage=True: Tuple of (result, TokenUsage) with token counts.
+
+        Raises:
+            OutputTooLongError: If output exceeds token limits.
+            Exception: Re-raises API errors after retries exhausted.
+        """
+        from anthropic import APIConnectionError, APIStatusError, RateLimitError
+
+        start_time = time.time()
+
+        # Convert OpenAI-style messages to Anthropic format
+        system_prompt = None
+        anthropic_messages = []
+
+        for msg in messages:
+            role = msg.get("role", "user")
+            content = msg.get("content", "")
+
+            if role == "system":
+                if system_prompt:
+                    system_prompt += "\n\n" + content
+                else:
+                    system_prompt = content
+            else:
+                anthropic_messages.append({"role": role, "content": content})
+
+        # Add JSON schema instruction if response_format is provided
+        if response_format is not None and hasattr(response_format, "model_json_schema"):
+            schema = response_format.model_json_schema()
+            schema_msg = f"\n\nYou must respond with valid JSON matching this schema:\n{json.dumps(schema, indent=2)}"
+            if system_prompt:
+                system_prompt += schema_msg
+            else:
+                system_prompt = schema_msg
+
+        # Prepare parameters
+        call_params: dict[str, Any] = {
+            "model": self.model,
+            "messages": anthropic_messages,
+            "max_tokens": max_completion_tokens if max_completion_tokens is not None else 4096,
+        }
+
+        if system_prompt:
+            call_params["system"] = system_prompt
+
+        if temperature is not None:
+            call_params["temperature"] = temperature
+
+        last_exception = None
+
+        for attempt in range(max_retries + 1):
+            try:
+                response = await self._client.messages.create(**call_params)
+
+                # Anthropic response content is a list of blocks
+                content = ""
+                for block in response.content:
+                    if block.type == "text":
+                        content += block.text
+
+                if response_format is not None:
+                    # Models may wrap JSON in markdown code blocks
+                    clean_content = content
+                    if "```json" in content:
+                        clean_content = content.split("```json")[1].split("```")[0].strip()
+                    elif "```" in content:
+                        clean_content = content.split("```")[1].split("```")[0].strip()
+
+                    try:
+                        json_data = json.loads(clean_content)
+                    except json.JSONDecodeError:
+                        # Fallback to parsing raw content if markdown stripping failed
+                        json_data = json.loads(content)
+
+                    if skip_validation:
+                        result = json_data
+                    else:
+                        result = response_format.model_validate(json_data)
+                else:
+                    result = content
+
+                # Record metrics and log slow calls
+                duration = time.time() - start_time
+                input_tokens = response.usage.input_tokens or 0 if response.usage else 0
+                output_tokens = response.usage.output_tokens or 0 if response.usage else 0
+                total_tokens = input_tokens + output_tokens
+
+                # Record LLM metrics
+                metrics = get_metrics_collector()
+                metrics.record_llm_call(
+                    provider=self.provider,
+                    model=self.model,
+                    scope=scope,
+                    duration=duration,
+                    input_tokens=input_tokens,
+                    output_tokens=output_tokens,
+                    success=True,
+                )
+
+                # Log slow calls
+                if duration > 10.0:
+                    logger.info(
+                        f"slow llm call: scope={scope}, model={self.provider}/{self.model}, "
+                        f"input_tokens={input_tokens}, output_tokens={output_tokens}, "
+                        f"time={duration:.3f}s"
+                    )
+
+                if return_usage:
+                    token_usage = TokenUsage(
+                        input_tokens=input_tokens,
+                        output_tokens=output_tokens,
+                        total_tokens=total_tokens,
+                    )
+                    return result, token_usage
+                return result
+
+            except json.JSONDecodeError as e:
+                last_exception = e
+                if attempt < max_retries:
+                    logger.warning("Anthropic returned invalid JSON, retrying...")
+                    backoff = min(initial_backoff * (2**attempt), max_backoff)
+                    await asyncio.sleep(backoff)
+                    continue
+                else:
+                    logger.error(f"Anthropic returned invalid JSON after {max_retries + 1} attempts")
+                    raise
+
+            except (APIConnectionError, RateLimitError, APIStatusError) as e:
+                # Fast fail on 401/403
+                if isinstance(e, APIStatusError) and e.status_code in (401, 403):
+                    logger.error(f"Anthropic auth error (HTTP {e.status_code}), not retrying: {str(e)}")
+                    raise
+
+                last_exception = e
+                if attempt < max_retries:
+                    # Check if it's a rate limit or server error
+                    should_retry = isinstance(e, (APIConnectionError, RateLimitError)) or (
+                        isinstance(e, APIStatusError) and e.status_code >= 500
+                    )
+
+                    if should_retry:
+                        backoff = min(initial_backoff * (2**attempt), max_backoff)
+                        jitter = backoff * 0.2 * (2 * (time.time() % 1) - 1)
+                        await asyncio.sleep(backoff + jitter)
+                        continue
+
+                logger.error(f"Anthropic API error after {max_retries + 1} attempts: {str(e)}")
+                raise
+
+            except Exception as e:
+                logger.error(f"Unexpected error during Anthropic call: {type(e).__name__}: {str(e)}")
+                raise
+
+        if last_exception:
+            raise last_exception
+        raise RuntimeError("Anthropic call failed after all retries")
+
+    async def call_with_tools(
+        self,
+        messages: list[dict[str, Any]],
+        tools: list[dict[str, Any]],
+        max_completion_tokens: int | None = None,
+        temperature: float | None = None,
+        scope: str = "tools",
+        max_retries: int = 5,
+        initial_backoff: float = 1.0,
+        max_backoff: float = 30.0,
+        tool_choice: str | dict[str, Any] = "auto",
+    ) -> LLMToolCallResult:
+        """
+        Make an LLM API call with tool/function calling support.
+
+        Args:
+            messages: List of message dicts. Can include tool results with role='tool'.
+            tools: List of tool definitions in OpenAI format.
+            max_completion_tokens: Maximum tokens in response.
+            temperature: Sampling temperature (0.0-2.0).
+            scope: Scope identifier for tracking.
+            max_retries: Maximum retry attempts.
+            initial_backoff: Initial backoff time in seconds.
+            max_backoff: Maximum backoff time in seconds.
+            tool_choice: How to choose tools - "auto", "none", "required", or specific function.
+
+        Returns:
+            LLMToolCallResult with content and/or tool_calls.
+        """
+        from anthropic import APIConnectionError, APIStatusError
+
+        start_time = time.time()
+
+        # Convert OpenAI tool format to Anthropic format
+        anthropic_tools = []
+        for tool in tools:
+            func = tool.get("function", {})
+            anthropic_tools.append(
+                {
+                    "name": func.get("name", ""),
+                    "description": func.get("description", ""),
+                    "input_schema": func.get("parameters", {"type": "object", "properties": {}}),
+                }
+            )
+
+        # Convert messages - handle tool results
+        system_prompt = None
+        anthropic_messages = []
+        for msg in messages:
+            role = msg.get("role", "user")
+            content = msg.get("content", "")
+
+            if role == "system":
+                system_prompt = (system_prompt + "\n\n" + content) if system_prompt else content
+            elif role == "tool":
+                # Anthropic uses tool_result blocks
+                anthropic_messages.append(
+                    {
+                        "role": "user",
+                        "content": [
+                            {"type": "tool_result", "tool_use_id": msg.get("tool_call_id", ""), "content": content}
+                        ],
+                    }
+                )
+            elif role == "assistant" and msg.get("tool_calls"):
+                # Convert assistant tool calls
+                tool_use_blocks = []
+                for tc in msg["tool_calls"]:
+                    tool_use_blocks.append(
+                        {
+                            "type": "tool_use",
+                            "id": tc.get("id", ""),
+                            "name": tc.get("function", {}).get("name", ""),
+                            "input": json.loads(tc.get("function", {}).get("arguments", "{}")),
+                        }
+                    )
+                anthropic_messages.append({"role": "assistant", "content": tool_use_blocks})
+            else:
+                anthropic_messages.append({"role": role, "content": content})
+
+        call_params: dict[str, Any] = {
+            "model": self.model,
+            "messages": anthropic_messages,
+            "tools": anthropic_tools,
+            "max_tokens": max_completion_tokens or 4096,
+        }
+        if system_prompt:
+            call_params["system"] = system_prompt
+
+        if temperature is not None:
+            call_params["temperature"] = temperature
+
+        last_exception = None
+        for attempt in range(max_retries + 1):
+            try:
+                response = await self._client.messages.create(**call_params)
+
+                # Extract content and tool calls
+                content_parts = []
+                tool_calls: list[LLMToolCall] = []
+
+                for block in response.content:
+                    if block.type == "text":
+                        content_parts.append(block.text)
+                    elif block.type == "tool_use":
+                        tool_calls.append(LLMToolCall(id=block.id, name=block.name, arguments=block.input or {}))
+
+                content = "".join(content_parts) if content_parts else None
+                finish_reason = "tool_calls" if tool_calls else "stop"
+
+                # Extract token usage
+                input_tokens = response.usage.input_tokens or 0
+                output_tokens = response.usage.output_tokens or 0
+
+                # Record metrics
+                metrics = get_metrics_collector()
+                metrics.record_llm_call(
+                    provider=self.provider,
+                    model=self.model,
+                    scope=scope,
+                    duration=time.time() - start_time,
+                    input_tokens=input_tokens,
+                    output_tokens=output_tokens,
+                    success=True,
+                )
+
+                return LLMToolCallResult(
+                    content=content,
+                    tool_calls=tool_calls,
+                    finish_reason=finish_reason,
+                    input_tokens=input_tokens,
+                    output_tokens=output_tokens,
+                )
+
+            except (APIConnectionError, APIStatusError) as e:
+                if isinstance(e, APIStatusError) and e.status_code in (401, 403):
+                    raise
+                last_exception = e
+                if attempt < max_retries:
+                    await asyncio.sleep(min(initial_backoff * (2**attempt), max_backoff))
+                    continue
+                raise
+
+        if last_exception:
+            raise last_exception
+        raise RuntimeError("Anthropic tool call failed")
+
+    async def cleanup(self) -> None:
+        """Clean up resources (close Anthropic client connections)."""
+        if hasattr(self, "_client") and self._client:
+            await self._client.close()