agentkernel 0.2.11__tar.gz → 0.2.13__tar.gz

{agentkernel-0.2.11 → agentkernel-0.2.13}/PKG-INFO

@@ -1,11 +1,11 @@
 Metadata-Version: 2.3
 Name: agentkernel
-Version: 0.2.11
+Version: 0.2.13
 Summary: Agent Kernel - Unified AI Agents Runtime
 Author: Yaala Labs
 Author-email: Yaala Labs <agentkernel@yaalalabs.com>
-License: MIT
-Requires-Dist: deprecated>=1.3.1
+License: Apache-2.0
+Requires-Dist: deprecated>=1.2.18
 Requires-Dist: pydantic>=2.11.7
 Requires-Dist: pydantic-settings>=2.10.1
 Requires-Dist: pyyaml>=6.0.2
@@ -17,7 +17,11 @@ Requires-Dist: openinference-instrumentation-google-adk>=0.1.6 ; extra == 'adk'
 Requires-Dist: fastapi>=0.118.0 ; extra == 'api'
 Requires-Dist: uvicorn>=0.37.0 ; extra == 'api'
 Requires-Dist: gunicorn>=23.0.0 ; extra == 'api'
+Requires-Dist: pyjwt>=2.0.0 ; extra == 'auth'
 Requires-Dist: boto3>=1.41.4 ; extra == 'aws'
+Requires-Dist: azure-functions>=1.24.0 ; extra == 'azure'
+Requires-Dist: azure-data-tables>=12.4.0 ; extra == 'azure'
+Requires-Dist: azure-core>=1.26.0 ; extra == 'azure'
 Requires-Dist: crewai>=0.150.0 ; extra == 'crewai'
 Requires-Dist: openinference-instrumentation-crewai>=0.1.16 ; extra == 'crewai'
 Requires-Dist: openinference-instrumentation-litellm>=0.1.28 ; extra == 'crewai'
@@ -57,7 +61,9 @@ Requires-Python: >=3.12, <3.14
 Provides-Extra: a2a
 Provides-Extra: adk
 Provides-Extra: api
+Provides-Extra: auth
 Provides-Extra: aws
+Provides-Extra: azure
 Provides-Extra: cli
 Provides-Extra: crewai
 Provides-Extra: gmail
@@ -80,15 +86,18 @@ Description-Content-Type: text/markdown
 [![PyPI version](https://badge.fury.io/py/agentkernel.svg)](https://badge.fury.io/py/agentkernel)
 [![Python 3.12+](https://img.shields.io/badge/python-3.12+-blue.svg)](https://www.python.org/downloads/)
 
-Agent Kernel is a lightweight runtime and adapter layer for building and running AI agents across multiple frameworks and running within a unified execution environment. Migrate your existing agents to Agent Kernel and instantly utilize pre-built execution and testing capabilities.
+Agent Kernel is a lightweight **multi-cloud AI agent runtime** and adapter layer for building and running AI agents across multiple frameworks and cloud providers. Deploy the same agent code to **AWS or Azure** without modification. Migrate your existing agents to Agent Kernel and instantly utilize pre-built execution and testing capabilities.
+
+**Supported Cloud Platforms:** AWS, Azure
 
 ## Features
 
 - **Unified API**: Common abstractions (Agent, Runner, Session, Module, Runtime) across frameworks
 - **Multi-Framework Support**: OpenAI Agents SDK, CrewAI, LangGraph, Google ADK
-- **Session Management**: Built-in session abstraction for conversational state
-- **Flexible Deployment**: Interactive CLI for local development and testing, AWS Lambda handler for serverless deployment, AWS ECS Fargate deployment
-- **Pluggable Architecture**: Easy to extend with custom framework adapters
+- **Multi-Cloud Deployment**: Deploy to AWS (Lambda, ECS/Fargate) or Azure (Functions, Container Apps) with the same code
+- **Session Management**: Built-in session abstraction with multi-cloud storage (Redis, DynamoDB, Cosmos DB)
+- **Flexible Deployment**: Interactive CLI, REST API, serverless (AWS Lambda, Azure Functions), containerized (AWS ECS, Azure Container Apps)
+- **Pluggable Architecture**: Easy to extend with custom framework adapters and cloud providers
 - **MCP Server**: Built-in Model Context Protocol server for exposing agents as MCP tools and exposing any custom tool
 - **A2A Server**: Built-in Agent-to-Agent communication server for exposing agents with a simple configuration change
 - **REST API**: Built-in REST API server for agent interaction
@@ -231,7 +240,11 @@ Then interact with your agents:
 (researcher) >> What is the latest news on AI?
 ```
 
-## AWS Lambda Deployment
+## Multi-Cloud Deployment
+
+Deploy your agents to AWS or Azure using the built-in cloud deployment handlers.
+
+### AWS Lambda Deployment
 
 Deploy your agents as serverless functions using the built-in Lambda handler.
 
@@ -248,6 +261,23 @@ OpenAIModule([assistant])
 handler = Lambda.handler
 ```
 
+### Azure Functions Deployment
+
+Deploy your agents as Azure Functions using the built-in Azure handler.
+
+```python
+from openai import OpenAI
+from agents import Agent as OpenAIAgent
+from agentkernel.azure import AzureFunctions
+from agentkernel.openai import OpenAIModule
+
+client = OpenAI()
+assistant = OpenAIAgent(name="assistant")
+
+OpenAIModule([assistant])
+handler = AzureFunctions.handler
+```
+
 **Request Format:**
 
 ```json
@@ -307,11 +337,11 @@ Supported formats: `.yaml`, `.yml`, `.json`
 
 #### Session Store
 
-Configure where agent sessions are stored.
+Configure where agent sessions are stored (supports multi-cloud storage backends).
 
 - **Field**: `session.type`
 - **Type**: string
-- **Options**: `in_memory`, `redis`
+- **Options**: `in_memory`, `redis`, `dynamodb` (AWS), `cosmosdb` (Azure)
 - **Default**: `in_memory`
 - **Environment Variable**: `AK_SESSION__TYPE`
 
@@ -1034,7 +1064,17 @@ The project uses:
 
 ## License
 
-MIT License - see LICENSE file for details.
+Unless otherwise specified, all content, including all source code files and documentation files in this repository are:
+
+Copyright (c) 2025-2026 Yaala Labs.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
+
+SPDX-License-Identifier: Apache-2.0
 
 ## Support
 

{agentkernel-0.2.11 → agentkernel-0.2.13}/README.md

@@ -3,15 +3,18 @@
 [![PyPI version](https://badge.fury.io/py/agentkernel.svg)](https://badge.fury.io/py/agentkernel)
 [![Python 3.12+](https://img.shields.io/badge/python-3.12+-blue.svg)](https://www.python.org/downloads/)
 
-Agent Kernel is a lightweight runtime and adapter layer for building and running AI agents across multiple frameworks and running within a unified execution environment. Migrate your existing agents to Agent Kernel and instantly utilize pre-built execution and testing capabilities.
+Agent Kernel is a lightweight **multi-cloud AI agent runtime** and adapter layer for building and running AI agents across multiple frameworks and cloud providers. Deploy the same agent code to **AWS or Azure** without modification. Migrate your existing agents to Agent Kernel and instantly utilize pre-built execution and testing capabilities.
+
+**Supported Cloud Platforms:** AWS, Azure
 
 ## Features
 
 - **Unified API**: Common abstractions (Agent, Runner, Session, Module, Runtime) across frameworks
 - **Multi-Framework Support**: OpenAI Agents SDK, CrewAI, LangGraph, Google ADK
-- **Session Management**: Built-in session abstraction for conversational state
-- **Flexible Deployment**: Interactive CLI for local development and testing, AWS Lambda handler for serverless deployment, AWS ECS Fargate deployment
-- **Pluggable Architecture**: Easy to extend with custom framework adapters
+- **Multi-Cloud Deployment**: Deploy to AWS (Lambda, ECS/Fargate) or Azure (Functions, Container Apps) with the same code
+- **Session Management**: Built-in session abstraction with multi-cloud storage (Redis, DynamoDB, Cosmos DB)
+- **Flexible Deployment**: Interactive CLI, REST API, serverless (AWS Lambda, Azure Functions), containerized (AWS ECS, Azure Container Apps)
+- **Pluggable Architecture**: Easy to extend with custom framework adapters and cloud providers
 - **MCP Server**: Built-in Model Context Protocol server for exposing agents as MCP tools and exposing any custom tool
 - **A2A Server**: Built-in Agent-to-Agent communication server for exposing agents with a simple configuration change
 - **REST API**: Built-in REST API server for agent interaction
@@ -154,7 +157,11 @@ Then interact with your agents:
 (researcher) >> What is the latest news on AI?
 ```
 
-## AWS Lambda Deployment
+## Multi-Cloud Deployment
+
+Deploy your agents to AWS or Azure using the built-in cloud deployment handlers.
+
+### AWS Lambda Deployment
 
 Deploy your agents as serverless functions using the built-in Lambda handler.
 
@@ -171,6 +178,23 @@ OpenAIModule([assistant])
 handler = Lambda.handler
 ```
 
+### Azure Functions Deployment
+
+Deploy your agents as Azure Functions using the built-in Azure handler.
+
+```python
+from openai import OpenAI
+from agents import Agent as OpenAIAgent
+from agentkernel.azure import AzureFunctions
+from agentkernel.openai import OpenAIModule
+
+client = OpenAI()
+assistant = OpenAIAgent(name="assistant")
+
+OpenAIModule([assistant])
+handler = AzureFunctions.handler
+```
+
 **Request Format:**
 
 ```json
@@ -230,11 +254,11 @@ Supported formats: `.yaml`, `.yml`, `.json`
 
 #### Session Store
 
-Configure where agent sessions are stored.
+Configure where agent sessions are stored (supports multi-cloud storage backends).
 
 - **Field**: `session.type`
 - **Type**: string
-- **Options**: `in_memory`, `redis`
+- **Options**: `in_memory`, `redis`, `dynamodb` (AWS), `cosmosdb` (Azure)
 - **Default**: `in_memory`
 - **Environment Variable**: `AK_SESSION__TYPE`
 
@@ -957,7 +981,17 @@ The project uses:
 
 ## License
 
-MIT License - see LICENSE file for details.
+Unless otherwise specified, all content, including all source code files and documentation files in this repository are:
+
+Copyright (c) 2025-2026 Yaala Labs.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
+
+SPDX-License-Identifier: Apache-2.0
 
 ## Support
 

{agentkernel-0.2.11 → agentkernel-0.2.13}/pyproject.toml

@@ -4,16 +4,16 @@ build-backend = "uv_build"
 
 [project]
 name = "agentkernel"
-version = "0.2.11"
+version = "0.2.13"
 description = "Agent Kernel - Unified AI Agents Runtime"
 readme = "README.md"
 requires-python = ">=3.12,<3.14"
-license = { text = "MIT" }
+license = { text = "Apache-2.0" }
 authors = [
     { name = "Yaala Labs", email = "agentkernel@yaalalabs.com" }
 ]
 dependencies = [
-    "deprecated>=1.3.1",
+    "deprecated>=1.2.18",
     "pydantic>=2.11.7",
     "pydantic-settings>=2.10.1",
     "pyyaml>=6.0.2",
@@ -21,6 +21,9 @@ dependencies = [
 ]
 
 [project.optional-dependencies]
+auth = [
+    "PyJWT>=2.0.0",
+]
 langfuse = [
     "langfuse>=3.9.2",
     "nest-asyncio>=1.6.0",
@@ -45,6 +48,12 @@ cli = [
 aws = [
     "boto3>=1.41.4",
 ]
+
+azure = [
+    "azure-functions>=1.24.0",
+    "azure-data-tables>=12.4.0",
+    "azure-core>=1.26.0"
+]
 redis = [
     "redis>=7.1.0",
 ]

{agentkernel-0.2.11 → agentkernel-0.2.13}/src/agentkernel/api/handler.py

@@ -27,16 +27,11 @@ class RESTRequestHandler(ABC):
         """
         Returns the APIRouter instance which has configured routes
         E.g.:
-        - GET /health: Health check
-        - GET /agents: List available agents
+        - GET /api/v1/agents: List available agents
 
         router = APIRouter()
 
-        @router.get("/health")
-        def health():
-            return {"status": "ok"}
-
-        @router.get("/agents")
+        @router.get("/api/v1/agents")
         def list_agents():
             from ..core.runtime import Runtime
             return {"agents": list(Runtime.current().agents().keys())}
@@ -49,9 +44,8 @@ class AgentRESTRequestHandler(RESTRequestHandler):
     """
     API routers that expose endpoints to interact with Agent Kernel.
     Endpoints:
-    - GET /health: Health check
-    - GET /agents: List available agents
-    - POST /run: Run an agent with a prompt
+    - GET /api/v1/agents: List available agents
+    - POST /api/v1/chat: Run an agent with a prompt
       Payload JSON: { "prompt": str, "agent": str | null, "session_id": str | null }
     """
 
@@ -89,19 +83,15 @@ class AgentRESTRequestHandler(RESTRequestHandler):
 
         router = APIRouter()
 
-        @router.get("/health")
-        def health():
-            return {"status": "ok"}
-
-        @router.get("/agents")
+        @router.get("/api/v1/agents")
         def list_agents():
             return {"agents": list(Runtime.current().agents().keys())}
 
-        @router.post("/run")
+        @router.post("/api/v1/chat")
         async def run(body: AgentRESTRequestHandler.RunRequest):
             return await self.run(body)
 
-        @router.post("/run-multipart")
+        @router.post("/api/v1/chat-multipart")
         async def run_multipart(
             prompt: str = Form(...),
             agent: Optional[str] = Form(None),
@@ -175,7 +165,7 @@ class AgentRESTRequestHandler(RESTRequestHandler):
         except HTTPException:
             raise
         except ValueError as e:
-            self._log.error(f"POST /run error: {e}\n{traceback.format_exc()}")
+            self._log.error(f"POST /api/v1/chat error: {e}\n{traceback.format_exc()}")
             raise HTTPException(
                 status_code=HTTPStatus.BAD_REQUEST,
                 detail={
@@ -184,7 +174,7 @@ class AgentRESTRequestHandler(RESTRequestHandler):
                 },
             )
         except Exception as e:
-            self._log.error(f"POST /run error: {e}\n{traceback.format_exc()}")
+            self._log.error(f"POST /api/v1/chat error: {e}\n{traceback.format_exc()}")
             raise HTTPException(
                 status_code=HTTPStatus.INTERNAL_SERVER_ERROR,
                 detail={
@@ -295,7 +285,7 @@ class AgentRESTRequestHandler(RESTRequestHandler):
         except HTTPException:
             raise
         except ValueError as e:
-            self._log.error(f"POST /run-multipart error: {e}\n{traceback.format_exc()}")
+            self._log.error(f"POST /api/v1/chat-multipart error: {e}\n{traceback.format_exc()}")
             raise HTTPException(
                 status_code=HTTPStatus.BAD_REQUEST,
                 detail={
@@ -304,7 +294,7 @@ class AgentRESTRequestHandler(RESTRequestHandler):
                 },
             )
         except Exception as e:
-            self._log.error(f"POST /run-multipart error: {e}\n{traceback.format_exc()}")
+            self._log.error(f"POST /api/v1/chat-multipart error: {e}\n{traceback.format_exc()}")
             raise HTTPException(
                 status_code=HTTPStatus.INTERNAL_SERVER_ERROR,
                 detail={

{agentkernel-0.2.11 → agentkernel-0.2.13}/src/agentkernel/api/http.py

@@ -1,10 +1,11 @@
 import logging
 
 import uvicorn
-from fastapi import APIRouter, FastAPI
+from fastapi import APIRouter, Depends, FastAPI, HTTPException, Request
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.openapi.utils import get_openapi
 
+from ..auth import AuthValidator, ValidationContext
 from ..core.config import AKConfig
 from .handler import AgentRESTRequestHandler, RESTRequestHandler
 
@@ -23,6 +24,14 @@ class RESTAPI:
 
     _log = logging.getLogger("ak.api.http")
     _custom_routers = []
+    _auth_token_validators = []
+
+    @classmethod
+    def _get_router_dependencies(cls):
+        """Get dependencies to apply to APIRouters.
+        :return: List of dependencies or None if no auth validators are configured
+        """
+        return cls._auth_token_validators if cls._auth_token_validators else None
 
     @classmethod
     def _create_app(cls, routers, lifespan=None) -> FastAPI:
@@ -30,6 +39,10 @@ class RESTAPI:
         Assembles a FastAPI app from routers.
         :param routers: List of routers to include in the app.
         :param lifespan: Optional lifespan handler.
+
+        Global endpoints:
+        - GET /health: Health check
+        - GET /openapi.json: OpenAPI specification
         """
         app = FastAPI(title="Agent Kernel REST API", debug=True, lifespan=lifespan)
 
@@ -41,8 +54,12 @@ class RESTAPI:
             allow_headers=["*"],
         )
 
+        @app.get("/health")
+        def health():
+            return {"status": "ok"}
+
         for r in routers or []:
-            app.include_router(r)
+            app.include_router(router=r, dependencies=cls._get_router_dependencies())
 
         @app.get("/openapi.json")
         async def get_openapi_endpoint():
@@ -56,6 +73,9 @@ class RESTAPI:
 
     @classmethod
     def add(cls, router: APIRouter):
+        """Add a custom router to the REST API.
+        :param router: FastAPI router to add to the API
+        """
         cls._log.debug(f"Adding custom router")
         for route in router.routes:
             cls._log.debug(f"Route: {route.path} [{route.methods}]")
@@ -63,8 +83,8 @@ class RESTAPI:
 
     @classmethod
     def run(cls, handlers: list[RESTRequestHandler] = None):
-        """
-        Starts the REST API server.
+        """Start the REST API server.
+        :param handlers: List of REST request handlers to use (default: AgentRESTRequestHandler)
         """
         if handlers is None:
             handlers = [AgentRESTRequestHandler()]
@@ -92,5 +112,36 @@ class RESTAPI:
             app = cls._create_app(routers=routers)
         # Add custom routers
         for router in cls._custom_routers:
-            app.include_router(router, prefix=AKConfig.get().api.custom_router_prefix)
+            app.include_router(router, prefix=AKConfig.get().api.custom_router_prefix, dependencies=cls._get_router_dependencies())
         uvicorn.run(app=app, host=host, port=port, reload=False)
+
+    @classmethod
+    def add_auth_handlers(cls, auth_validators: list[AuthValidator]):
+        """Add authentication validators to the REST API.
+        :param auth_validators: List of auth validators to add for token validation
+        """
+
+        def get_auth_function(token_validator: AuthValidator):
+            def verify_token(request: Request):
+                """Verify authentication token for incoming requests.
+                :param request: FastAPI Request object
+                :return: ValidationResult if token is valid
+                :raises: HTTPException if authentication fails
+                """
+                request_url = str(request.url)
+                cls._log.debug(f"Validating token for request: {request.method} {request_url}")
+                auth_token = request.headers.get("authorization")
+                if auth_token is None:
+                    raise HTTPException(status_code=401, detail="Missing authorization header")
+                auth_token = auth_token.replace("Bearer ", "").strip()
+                result = token_validator.validate(
+                    token=auth_token, context=ValidationContext(path=request_url, http_method=request.method, headers=dict(request.headers))
+                )
+                if not result.is_valid:
+                    raise HTTPException(status_code=401, detail=result.error_msg or "Unauthorized")
+                return result
+
+            return verify_token
+
+        for token_validator in auth_validators:
+            cls._auth_token_validators.append(Depends(get_auth_function(token_validator)))

agentkernel-0.2.13/src/agentkernel/auth/__init__.py

@@ -0,0 +1,14 @@
+"""
+Agent Kernel Auth package.
+
+This package contains the Auth implementation for exposing Agents.
+"""
+
+import importlib.metadata
+
+try:
+    __version__ = importlib.metadata.version("agentkernel")
+except importlib.metadata.PackageNotFoundError:
+    __version__ = "0.1.0"
+
+from .handler import AuthValidator, ValidationContext, ValidationResult

agentkernel-0.2.13/src/agentkernel/auth/handler.py

@@ -0,0 +1,69 @@
+import base64
+import hashlib
+import hmac
+from abc import ABC, abstractmethod
+from typing import Any, Dict, Optional
+
+import jwt
+from pydantic import BaseModel
+
+
+class ValidationContext(BaseModel):
+    path: Optional[str] = None
+    http_method: Optional[str] = None
+    headers: Dict[str, str]
+
+
+class ValidationResult(BaseModel):
+    is_valid: bool
+    subject: Optional[str] = "user"
+    claims: Optional[Dict[str, Any]] = None
+    error_msg: Optional[str] = None
+
+
+class AuthValidator(ABC):
+    """Base class for token validation.
+    The validate() method must be implemented by subclasses,
+    There are some basic built-in cryptographic validation helpers which can be used if needed."""
+
+    @abstractmethod
+    def validate(self, token: str, context: Optional[ValidationContext] = None) -> ValidationResult:
+        """Main validation entry point. This is where the custom logic has to be written.
+        :param token: The authentication token to validate
+        :param context: Optional validation context containing request metadata
+        :return: ValidationResult indicating if token is valid and associated claims
+        """
+        pass
+
+    def _validate_hmac(self, message: bytes, signature: str, secret: str, algorithm: str = "sha256") -> bool:
+        """Basic HMAC signature validation.
+        :param message: The message bytes to verify
+        :param signature: The HMAC signature to validate against
+        :param secret: The secret key used for HMAC generation
+        :param algorithm: Hash algorithm to use (default: sha256)
+        :return: True if signature is valid, False otherwise
+        """
+        mac = hmac.new(key=secret.encode(), msg=message, digestmod=getattr(hashlib, algorithm))
+        expected = base64.b64encode(mac.digest()).decode()
+        return hmac.compare_digest(expected, signature)
+
+    def _validate_rs256_jwt(
+        self, token: str, public_key: str, audience: Optional[str] = None, issuer: Optional[str] = None, options: Optional[Dict[str, Any]] = None
+    ) -> Dict[str, Any]:
+        """Basic RS256 JWT validation.
+        :param token: The JWT token to validate
+        :param public_key: The RSA public key for signature verification
+        :param audience: Optional expected audience claim
+        :param issuer: Optional expected issuer claim
+        :param options: Additional JWT validation options
+        :return: Decoded JWT claims dictionary
+        :raises: jwt exceptions if token is invalid
+        """
+        return jwt.decode(
+            token,
+            public_key,
+            algorithms=["RS256"],
+            audience=audience,
+            issuer=issuer,
+            options=options or {},
+        )

agentkernel-0.2.13/src/agentkernel/azure.py

@@ -0,0 +1,8 @@
+import importlib.metadata
+
+try:
+    __version__ = importlib.metadata.version("agentkernel")
+except importlib.metadata.PackageNotFoundError:
+    __version__ = "0.1.0"
+
+from .deployment.azure import *

{agentkernel-0.2.11 → agentkernel-0.2.13}/src/agentkernel/core/__init__.py

@@ -27,4 +27,5 @@ from .module import Module
 from .runtime import Runtime, AuxiliaryCache
 from .service import AgentService
 from .hooks import PreHook, PostHook
+from .tool import ToolContext, ToolBuilder
 from .util.key_value_cache import KeyValueCache