alita-sdk 0.3.376__py3-none-any.whl → 0.3.423__py3-none-any.whl

alita_sdk/runtime/tools/sandbox.py

@@ -2,21 +2,60 @@ import asyncio
 import logging
 import subprocess
 import os
-from typing import Any, Type, Optional, Dict
-from langchain_core.tools import BaseTool
-from pydantic import BaseModel, create_model
+from typing import Any, Type, Optional, Dict, List, Literal, Union
+from copy import deepcopy
+from pathlib import Path
+
+from langchain_core.tools import BaseTool, BaseToolkit
+from langchain_core.messages import ToolCall
+from pydantic import BaseModel, create_model, ConfigDict, Field
 from pydantic.fields import FieldInfo
 
 logger = logging.getLogger(__name__)
 
+name = "pyodide"
+
+
+def get_tools(tools_list: list, alita_client=None, llm=None, memory_store=None):
+    """
+    Get sandbox tools for the provided tool configurations.
+
+    Args:
+        tools_list: List of tool configurations
+        alita_client: Alita client instance for sandbox tools
+        llm: LLM client instance (unused for sandbox)
+        memory_store: Optional memory store instance (unused for sandbox)
+
+    Returns:
+        List of sandbox tools
+    """
+    all_tools = []
+
+    for tool in tools_list:
+        if tool.get('type') == 'sandbox' or tool.get('toolkit_name') == 'sandbox':
+            try:
+                toolkit_instance = SandboxToolkit.get_toolkit(
+                    stateful=tool['settings'].get('stateful', False),
+                    allow_net=tool['settings'].get('allow_net', True),
+                    alita_client=alita_client,
+                    toolkit_name=tool.get('toolkit_name', '')
+                )
+                all_tools.extend(toolkit_instance.get_tools())
+            except Exception as e:
+                logger.error(f"Error in sandbox toolkit get_tools: {e}")
+                logger.error(f"Tool config: {tool}")
+                raise
+
+    return all_tools
+
 
 def _is_deno_available() -> bool:
     """Check if Deno is available in the PATH"""
     try:
         result = subprocess.run(
-            ["deno", "--version"], 
-            capture_output=True, 
-            text=True, 
+            ["deno", "--version"],
+            capture_output=True,
+            text=True,
             timeout=10
         )
         return result.returncode == 0
@@ -25,43 +64,17 @@ def _is_deno_available() -> bool:
 
 
 def _setup_pyodide_cache_env() -> None:
-    """Setup Pyodide caching environment variables for performance optimization"""
+    """Setup Pyodide caching environment variables for performance optimization [NO-OP]"""
     try:
-        # Check if cache environment file exists and source it
-        cache_env_file = os.path.expanduser("~/.pyodide_cache_env")
-        if os.path.exists(cache_env_file):
-            with open(cache_env_file, 'r') as f:
-                for line in f:
-                    line = line.strip()
-                    if line.startswith('export ') and '=' in line:
-                        # Parse export VAR=value format
-                        var_assignment = line[7:]  # Remove 'export '
-                        if '=' in var_assignment:
-                            key, value = var_assignment.split('=', 1)
-                            # Remove quotes if present
-                            value = value.strip('"').strip("'")
-                            os.environ[key] = value
-                            logger.debug(f"Set Pyodide cache env: {key}={value}")
-        
-        # Set default caching environment variables if not already set
-        cache_defaults = {
-            'PYODIDE_PACKAGES_PATH': os.path.expanduser('~/.cache/pyodide'),
-            'DENO_DIR': os.path.expanduser('~/.cache/deno'),
-            'PYODIDE_CACHE_DIR': os.path.expanduser('~/.cache/pyodide'),
-        }
-        
-        for key, default_value in cache_defaults.items():
-            if key not in os.environ:
-                os.environ[key] = default_value
-                logger.debug(f"Set default Pyodide env: {key}={default_value}")
-                
+        for key in ["SANDBOX_BASE", "DENO_DIR"]:
+            logger.info("Sandbox env: %s -> %s", key, os.environ.get(key, "n/a"))
     except Exception as e:
         logger.warning(f"Could not setup Pyodide cache environment: {e}")
 
 
 # Create input schema for the sandbox tool
 sandbox_tool_input = create_model(
-    "SandboxToolInput", 
+    "SandboxToolInput",
     code=(str, FieldInfo(description="Python code to execute in the sandbox environment"))
 )
 
@@ -72,7 +85,7 @@ class PyodideSandboxTool(BaseTool):
     This tool leverages langchain-sandbox to provide a safe environment for running untrusted Python code.
     Optimized for performance with caching and stateless execution by default.
     """
-    
+
     name: str = "pyodide_sandbox"
     description: str = """Execute Python code in a secure sandbox environment using Pyodide.
     This tool allows safe execution of Python code without access to the host system.
@@ -81,7 +94,7 @@ class PyodideSandboxTool(BaseTool):
     - Perform calculations or data analysis
     - Test Python algorithms
     - Run code that requires isolation from the host system
-    
+
     The sandbox supports most Python standard library modules and can install additional packages.
     Note: File access and some system operations are restricted for security.
     Optimized for performance with local caching (stateless by default for faster execution).
@@ -91,14 +104,37 @@ class PyodideSandboxTool(BaseTool):
     allow_net: bool = True
     session_bytes: Optional[bytes] = None
     session_metadata: Optional[Dict] = None
-    
+    alita_client: Optional[Any] = None
+
     def __init__(self, **kwargs: Any) -> None:
         super().__init__(**kwargs)
         self._sandbox = None
         # Setup caching environment for optimal performance
         _setup_pyodide_cache_env()
         self._initialize_sandbox()
-    
+
+    def _prepare_pyodide_input(self, code: str) -> str:
+        """Prepare input for PyodideSandboxTool by injecting state and alita_client into the code block."""
+        pyodide_predata = ""
+
+        # Add alita_client if available
+        if self.alita_client:
+            try:
+                # Get the directory of the current file and construct the path to sandbox_client.py
+                current_dir = Path(__file__).parent
+                sandbox_client_path = current_dir.parent / 'clients' / 'sandbox_client.py'
+
+                with open(sandbox_client_path, 'r') as f:
+                    sandbox_client_code = f.read()
+                pyodide_predata += f"{sandbox_client_code}\n"
+                pyodide_predata += (f"alita_client = SandboxClient(base_url='{self.alita_client.base_url}',"
+                                    f"project_id={self.alita_client.project_id},"
+                                    f"auth_token='{self.alita_client.auth_token}')\n")
+            except FileNotFoundError:
+                logger.error(f"sandbox_client.py not found. Ensure the file exists.")
+
+        return f"#elitea simplified client\n{pyodide_predata}{code}"
+
     def _initialize_sandbox(self) -> None:
         """Initialize the PyodideSandbox instance with optimized settings"""
         try:
@@ -110,12 +146,22 @@ class PyodideSandboxTool(BaseTool):
                 )
                 logger.error(error_msg)
                 raise RuntimeError(error_msg)
-            
+
             from langchain_sandbox import PyodideSandbox
-            
+
+            # Air-gapped settings
+            sandbox_base = os.environ.get("SANDBOX_BASE", os.path.expanduser('~/.cache/pyodide'))
+            sandbox_tmp = os.path.join(sandbox_base, "tmp")
+            deno_cache = os.environ.get("DENO_DIR", os.path.expanduser('~/.cache/deno'))
+
             # Configure sandbox with performance optimizations
             self._sandbox = PyodideSandbox(
                 stateful=self.stateful,
+                #
+                allow_env=["SANDBOX_BASE"],
+                allow_read=[sandbox_base, sandbox_tmp, deno_cache],
+                allow_write=[sandbox_tmp, deno_cache],
+                #
                 allow_net=self.allow_net,
                 # Use auto node_modules_dir for better caching
                 node_modules_dir="auto"
@@ -135,7 +181,7 @@ class PyodideSandboxTool(BaseTool):
         except Exception as e:
             logger.error(f"Failed to initialize PyodideSandbox: {e}")
             raise
-    
+
     def _run(self, code: str) -> str:
         """
         Synchronous version - runs the async method in a new event loop
@@ -144,7 +190,10 @@ class PyodideSandboxTool(BaseTool):
             # Check if sandbox is initialized, if not try to initialize
             if self._sandbox is None:
                 self._initialize_sandbox()
-            
+
+            # Prepare code with state and client injection
+            prepared_code = self._prepare_pyodide_input(code)
+
             # Check if we're already in an async context
             try:
                 loop = asyncio.get_running_loop()
@@ -152,11 +201,11 @@ class PyodideSandboxTool(BaseTool):
                 # We'll need to use a different approach
                 import concurrent.futures
                 with concurrent.futures.ThreadPoolExecutor() as executor:
-                    future = executor.submit(asyncio.run, self._arun(code))
+                    future = executor.submit(asyncio.run, self._arun(prepared_code))
                     return future.result()
             except RuntimeError:
                 # No running loop, safe to use asyncio.run
-                return asyncio.run(self._arun(code))
+                return asyncio.run(self._arun(prepared_code))
         except (ImportError, RuntimeError) as e:
             # Handle specific dependency errors gracefully
             error_msg = str(e)
@@ -169,7 +218,7 @@ class PyodideSandboxTool(BaseTool):
         except Exception as e:
             logger.error(f"Error executing code in sandbox: {e}")
             return f"Error executing code: {str(e)}"
-    
+
     async def _arun(self, code: str) -> str:
         """
         Execute Python code in the Pyodide sandbox
@@ -177,19 +226,19 @@ class PyodideSandboxTool(BaseTool):
         try:
             if self._sandbox is None:
                 self._initialize_sandbox()
-            
+
             # Execute the code with session state if available
             result = await self._sandbox.execute(
                 code,
                 session_bytes=self.session_bytes,
                 session_metadata=self.session_metadata
             )
-            
+
             # Update session state for stateful execution
             if self.stateful:
                 self.session_bytes = result.session_bytes
                 self.session_metadata = result.session_metadata
-            
+
             result_dict = {}
 
             if result.result is not None:
@@ -212,10 +261,10 @@ class PyodideSandboxTool(BaseTool):
 
             result_dict["execution_info"] = execution_info
             return result_dict
-            
+
         except Exception as e:
             logger.error(f"Error executing code in sandbox: {e}")
-            return f"Error executing code: {str(e)}"
+            return {"error": f"Error executing code: {str(e)}"}
 
 
 class StatefulPyodideSandboxTool(PyodideSandboxTool):
@@ -223,7 +272,7 @@ class StatefulPyodideSandboxTool(PyodideSandboxTool):
     A stateful version of the PyodideSandboxTool that maintains state between executions.
     This version preserves variables, imports, and function definitions across multiple tool calls.
     """
-    
+
     name: str = "stateful_pyodide_sandbox"
     description: str = """Execute Python code in a stateful sandbox environment using Pyodide.
     This tool maintains state between executions, preserving variables, imports, and function definitions.
@@ -232,41 +281,95 @@ class StatefulPyodideSandboxTool(PyodideSandboxTool):
     - Maintain variables across multiple calls
     - Develop complex programs step by step
     - Preserve imported libraries and defined functions
-    
+
     The sandbox supports most Python standard library modules and can install additional packages.
     Note: File access and some system operations are restricted for security.
     """
-    
+
     def __init__(self, **kwargs: Any) -> None:
         kwargs['stateful'] = True  # Force stateful mode
         super().__init__(**kwargs)
 
 
 # Factory function for creating sandbox tools
-def create_sandbox_tool(stateful: bool = False, allow_net: bool = True) -> BaseTool:
+def create_sandbox_tool(stateful: bool = False, allow_net: bool = True, alita_client: Optional[Any] = None) -> BaseTool:
     """
     Factory function to create sandbox tools with specified configuration.
-    
+
     Note: This tool requires Deno to be installed and available in PATH.
     For installation and optimization, run the bootstrap.sh script.
-    
+
     Args:
         stateful: Whether to maintain state between executions (default: False for better performance)
         allow_net: Whether to allow network access (for package installation)
-    
+
     Returns:
         Configured sandbox tool instance
-        
+
     Raises:
         ImportError: If langchain-sandbox is not installed
         RuntimeError: If Deno is not found in PATH
-        
+
     Performance Notes:
         - Stateless mode (default) is faster and avoids session state overhead
         - Run bootstrap.sh script to enable local caching and reduce initialization time
         - Cached wheels reduce package download time from ~4.76s to near-instant
     """
     if stateful:
-        return StatefulPyodideSandboxTool(allow_net=allow_net)
+        return StatefulPyodideSandboxTool(allow_net=allow_net, alita_client=alita_client)
     else:
-        return PyodideSandboxTool(stateful=False, allow_net=allow_net)
+        return PyodideSandboxTool(stateful=False, allow_net=allow_net, alita_client=alita_client)
+
+
+class SandboxToolkit(BaseToolkit):
+    tools: List[BaseTool] = []
+
+    @staticmethod
+    def toolkit_config_schema() -> Type[BaseModel]:
+        # Create sample tools to get their schemas
+        sample_tools = [
+            PyodideSandboxTool(),
+            StatefulPyodideSandboxTool()
+        ]
+        selected_tools = {x.name: x.args_schema.model_json_schema() for x in sample_tools}
+
+        return create_model(
+            'sandbox',
+            stateful=(bool, Field(default=False, description="Whether to maintain state between executions")),
+            allow_net=(bool, Field(default=True, description="Whether to allow network access for package installation")),
+            selected_tools=(List[Literal[tuple(selected_tools)]],
+                            Field(default=[], json_schema_extra={'args_schemas': selected_tools})),
+
+            __config__=ConfigDict(json_schema_extra={
+                'metadata': {
+                    "label": "Python Sandbox",
+                    "icon_url": "sandbox.svg",
+                    "hidden": False,
+                    "categories": ["code", "execution", "internal_tool"],
+                    "extra_categories": ["python", "pyodide", "sandbox", "code execution"],
+                }
+            })
+        )
+
+    @classmethod
+    def get_toolkit(cls, stateful: bool = False, allow_net: bool = True, alita_client=None, **kwargs):
+        """
+        Get toolkit with sandbox tools.
+
+        Args:
+            stateful: Whether to maintain state between executions
+            allow_net: Whether to allow network access
+            alita_client: Alita client instance for sandbox tools
+            **kwargs: Additional arguments
+        """
+        tools = []
+
+        if stateful:
+            tools.append(StatefulPyodideSandboxTool(allow_net=allow_net, alita_client=alita_client))
+        else:
+            tools.append(PyodideSandboxTool(stateful=False, allow_net=allow_net, alita_client=alita_client))
+
+        return cls(tools=tools)
+
+    def get_tools(self):
+        return self.tools

alita_sdk/runtime/tools/vectorstore.py

@@ -215,7 +215,7 @@ class VectorStoreWrapper(BaseToolApiWrapper):
         """List all collections in the vectorstore.
         Returns a list of collection names, or if no collections exist,
         returns a dict with an empty list and a message."""
-        raw = self.vector_adapter.list_indexes(self)
+        raw = self.vector_adapter.list_collections(self)
         # Normalize raw result to a list of names
         if not raw:
             # No collections found
@@ -414,7 +414,8 @@ class VectorStoreWrapper(BaseToolApiWrapper):
                 return {"status": "error", "message": f"Error: {format_exc()}"}
         if _documents:
             add_documents(vectorstore=self.vectorstore, documents=_documents)
-        return {"status": "ok", "message": f"successfully indexed {documents_count} documents"}
+        return {"status": "ok", "message": f"successfully indexed {documents_count} documents" if documents_count > 0
+        else "No new documents to index."}
 
     def search_documents(self, query:str, doctype: str = 'code', 
                          filter:dict|str={}, cut_off: float=0.5,

alita_sdk/runtime/tools/vectorstore_base.py

@@ -208,10 +208,10 @@ class VectorStoreWrapperBase(BaseToolApiWrapper):
             logger.error(f"Error during similarity search: {str(e)}")
             raise ToolException(f"Search failed: {str(e)}")
 
-    def list_indexes(self) -> List[str]:
+    def list_collections(self) -> List[str]:
         """List all collections in the vectorstore."""
 
-        collections = self.vector_adapter.list_indexes(self)
+        collections = self.vector_adapter.list_collections(self)
         if not collections:
             return "No indexed collections"
         return collections
@@ -308,7 +308,8 @@ class VectorStoreWrapperBase(BaseToolApiWrapper):
                 return {"status": "error", "message": f"Error: {format_exc()}"}
         if _documents:
             add_documents(vectorstore=self.vectorstore, documents=_documents)
-        return {"status": "ok", "message": f"successfully indexed {documents_count} documents"}
+        return {"status": "ok", "message": f"successfully indexed {documents_count} documents" if documents_count > 0
+        else "no documents to index"}
 
     def search_documents(self, query:str, doctype: str = 'code', 
                          filter:dict|str={}, cut_off: float=0.5,

alita_sdk/runtime/utils/utils.py

@@ -14,6 +14,7 @@ class IndexerKeywords(Enum):
     INDEX_META_TYPE = 'index_meta'
     INDEX_META_IN_PROGRESS = 'in_progress'
     INDEX_META_COMPLETED = 'completed'
+    INDEX_META_FAILED = 'failed'
 
 # This pattern matches characters that are NOT alphanumeric, underscores, or hyphens
 clean_string_pattern = re.compile(r'[^a-zA-Z0-9_.-]')

alita_sdk/tools/__init__.py

@@ -90,62 +90,74 @@ available_count = len(AVAILABLE_TOOLS)
 total_attempted = len(AVAILABLE_TOOLS) + len(FAILED_IMPORTS)
 logger.info(f"Tool imports completed: {available_count}/{total_attempted} successful")
 
+
 def get_tools(tools_list, alita, llm, store: Optional[BaseStore] = None, *args, **kwargs):
     tools = []
+
     for tool in tools_list:
-        # validate tool name syntax - it cannot be started with _
-        for tool_name in tool.get('settings', {}).get('selected_tools', []):
-            if isinstance(tool_name, str) and tool_name.startswith('_'):
-                raise ValueError(f"Tool name '{tool_name}' from toolkit '{tool.get('type', '')}' cannot start with '_'")
-
-        tool['settings']['alita'] = alita
-        tool['settings']['llm'] = llm
-        tool['settings']['store'] = store
+        settings = tool.get('settings')
+
+        # Skip tools without settings early
+        if not settings:
+            logger.warning(f"Tool '{tool.get('type', '')}' has no settings, skipping...")
+            continue
+
+        # Validate tool names once
+        selected_tools = settings.get('selected_tools', [])
+        invalid_tools = [name for name in selected_tools if isinstance(name, str) and name.startswith('_')]
+        if invalid_tools:
+            raise ValueError(f"Tool names {invalid_tools} from toolkit '{tool.get('type', '')}' cannot start with '_'")
+
+        # Cache tool type and add common settings
         tool_type = tool['type']
+        settings['alita'] = alita
+        settings['llm'] = llm
+        settings['store'] = store
+
+        # Set pgvector collection schema if present
+        if settings.get('pgvector_configuration'):
+            settings['pgvector_configuration']['collection_schema'] = str(tool['id'])
 
-        # Handle special cases for ADO tools
+        # Handle ADO special cases
         if tool_type in ['ado_boards', 'ado_wiki', 'ado_plans']:
             tools.extend(AVAILABLE_TOOLS['ado']['get_tools'](tool_type, tool))
+            continue
 
-        # Check if tool is available and has get_tools function
-        elif tool_type in AVAILABLE_TOOLS and 'get_tools' in AVAILABLE_TOOLS[tool_type]:
+        # Handle ADO repos aliases
+        if tool_type in ['ado_repos', 'azure_devops_repos'] and 'ado_repos' in AVAILABLE_TOOLS:
             try:
-                get_tools_func = AVAILABLE_TOOLS[tool_type]['get_tools']
-                tools.extend(get_tools_func(tool))
-
+                tools.extend(AVAILABLE_TOOLS['ado_repos']['get_tools'](tool))
             except Exception as e:
-                logger.error(f"Error getting tools for {tool_type}: {e}")
-                raise ToolException(f"Error getting tools for {tool_type}: {e}")
+                logger.error(f"Error getting ADO repos tools: {e}")
+            continue
 
-        # Handle ADO repos special case (it might be requested as azure_devops_repos)
-        elif tool_type in ['ado_repos', 'azure_devops_repos'] and 'ado_repos' in AVAILABLE_TOOLS:
+        # Handle standard tools
+        if tool_type in AVAILABLE_TOOLS and 'get_tools' in AVAILABLE_TOOLS[tool_type]:
             try:
-                get_tools_func = AVAILABLE_TOOLS['ado_repos']['get_tools']
-                tools.extend(get_tools_func(tool))
+                tools.extend(AVAILABLE_TOOLS[tool_type]['get_tools'](tool))
             except Exception as e:
-                logger.error(f"Error getting ADO repos tools: {e}")
+                logger.error(f"Error getting tools for {tool_type}: {e}")
+                raise ToolException(f"Error getting tools for {tool_type}: {e}")
+            continue
 
         # Handle custom modules
-        elif tool.get("settings", {}).get("module"):
+        if settings.get("module"):
             try:
-                settings = tool.get("settings", {})
                 mod = import_module(settings.pop("module"))
                 tkitclass = getattr(mod, settings.pop("class"))
-                #
-                get_toolkit_params = tool["settings"].copy()
+                get_toolkit_params = settings.copy()
                 get_toolkit_params["name"] = tool.get("name")
-                #
                 toolkit = tkitclass.get_toolkit(**get_toolkit_params)
                 tools.extend(toolkit.get_tools())
             except Exception as e:
                 logger.error(f"Error in getting custom toolkit: {e}")
+            continue
 
+        # Tool not available
+        if tool_type in FAILED_IMPORTS:
+            logger.warning(f"Tool '{tool_type}' is not available: {FAILED_IMPORTS[tool_type]}")
         else:
-            # Tool not available or not found
-            if tool_type in FAILED_IMPORTS:
-                logger.warning(f"Tool '{tool_type}' is not available: {FAILED_IMPORTS[tool_type]}")
-            else:
-                logger.warning(f"Unknown tool type: {tool_type}")
+            logger.warning(f"Unknown tool type: {tool_type}")
 
     return tools
 

alita_sdk/tools/ado/work_item/ado_wrapper.py

@@ -329,11 +329,14 @@ class AzureDevOpsApiWrapper(NonCodeIndexerToolkit):
                 parsed_item.update(fields_data)
 
             # extract relations if any
-            relations_data = work_item.relations
+            relations_data = None
+            if expand and str(expand).lower() in ("relations", "all"):
+                try:
+                    relations_data = getattr(work_item, 'relations', None)
+                except KeyError:
+                    relations_data = None
             if relations_data:
-                parsed_item['relations'] = []
-                for relation in relations_data:
-                    parsed_item['relations'].append(relation.as_dict())
+                parsed_item['relations'] = [relation.as_dict() for relation in relations_data]
 
             if parse_attachments:
                 # describe images in work item fields if present
@@ -344,13 +347,19 @@ class AzureDevOpsApiWrapper(NonCodeIndexerToolkit):
                         for img in images:
                             src = img.get('src')
                             if src:
-                                description = self.parse_attachment_by_url(src, image_description_prompt)
+                                description = self.parse_attachment_by_url(src, image_description_prompt=image_description_prompt)
                                 img['image-description'] = description
                         parsed_item[field_name] = str(soup)
                 # parse attached documents if present
-                if parsed_item['relations']:
-                    for attachment in parsed_item['relations']:
-                        attachment['content'] = self.parse_attachment_by_url(attachment['url'], attachment['attributes']['name'], image_description_prompt)
+                for relation in parsed_item.get('relations', []):
+                    # Only process actual file attachments
+                    if relation.get('rel') == 'AttachedFile':
+                        file_name = relation.get('attributes', {}).get('name')
+                        if file_name:
+                            try:
+                                relation['content'] = self.parse_attachment_by_url(relation['url'], file_name, image_description_prompt=image_description_prompt)
+                            except Exception as att_e:
+                                logger.warning(f"Failed to parse attachment {file_name}: {att_e}")
 
 
             return parsed_item