alita-sdk 0.3.257__py3-none-any.whl → 0.3.584__py3-none-any.whl

alita_sdk/runtime/utils/toolkit_utils.py

@@ -12,7 +12,9 @@ logger = logging.getLogger(__name__)
 
 def instantiate_toolkit_with_client(toolkit_config: Dict[str, Any], 
                                    llm_client: Any, 
-                                   alita_client: Optional[Any] = None) -> List[Any]:
+                                   alita_client: Optional[Any] = None,
+                                   mcp_tokens: Optional[Dict[str, Any]] = None,
+                                   use_prefix: bool = False) -> List[Any]:
     """
     Instantiate a toolkit with LLM client support.
     
@@ -22,20 +24,25 @@ def instantiate_toolkit_with_client(toolkit_config: Dict[str, Any],
     Args:
         toolkit_config: Configuration dictionary for the toolkit
         llm_client: LLM client instance for tools that need LLM capabilities
-        client: Optional additional client instance
+        alita_client: Optional additional client instance
+        mcp_tokens: Optional dictionary of MCP OAuth tokens by server URL
+        use_prefix: If True, tools get prefixed with toolkit_name to prevent collisions
+                   (for agent use). If False, tools use base names only (for testing interface).
+                   Default False for backward compatibility with testing.
     
     Returns:
         List of instantiated tools from the toolkit
         
     Raises:
         ValueError: If required configuration or client is missing
+        McpAuthorizationRequired: If MCP server requires OAuth authorization
         Exception: If toolkit instantiation fails
     """
+    toolkit_name = toolkit_config.get('toolkit_name', 'unknown')
     try:
         from ..toolkits.tools import get_tools
         
-        toolkit_name = toolkit_config.get('toolkit_name')
-        if not toolkit_name:
+        if not toolkit_name or toolkit_name == 'unknown':
             raise ValueError("toolkit_name is required in configuration")
         
         if not llm_client:
@@ -46,18 +53,22 @@ def instantiate_toolkit_with_client(toolkit_config: Dict[str, Any],
         # Log the configuration being used
         logger.info(f"Instantiating toolkit {toolkit_name} with LLM client")
         logger.debug(f"Toolkit {toolkit_name} configuration: {toolkit_config}")
-        
+
+        # Use toolkit type from config, or fall back to lowercase toolkit name
+        toolkit_type = toolkit_config.get('type', toolkit_name.lower())
+
         # Create a tool configuration dict with required fields
+        # Note: MCP toolkit always requires toolkit_name, other toolkits respect use_prefix flag
         tool_config = {
             'id': toolkit_config.get('id', random.randint(1, 1000000)),
-            'type': toolkit_config.get('type', toolkit_name.lower()),
+            'type': toolkit_config.get('type', toolkit_type),
             'settings': settings,
-            'toolkit_name': toolkit_name
+            'toolkit_name': toolkit_name if (use_prefix or toolkit_type == 'mcp') else None
         }
         
         # Get tools using the toolkit configuration with clients
-        # Parameter order: get_tools(tools_list, alita_client, llm, memory_store)
-        tools = get_tools([tool_config], alita_client, llm_client)
+        # Parameter order: get_tools(tools_list, alita_client, llm, memory_store, debug_mode, mcp_tokens)
+        tools = get_tools([tool_config], alita_client, llm_client, mcp_tokens=mcp_tokens)
         
         if not tools:
             logger.warning(f"No tools returned for toolkit {toolkit_name}")
@@ -67,6 +78,14 @@ def instantiate_toolkit_with_client(toolkit_config: Dict[str, Any],
         return tools
             
     except Exception as e:
+        # Re-raise McpAuthorizationRequired without logging as error
+        from ..utils.mcp_oauth import McpAuthorizationRequired
+        
+        if isinstance(e, McpAuthorizationRequired):
+            logger.info(f"Toolkit {toolkit_name} requires MCP OAuth authorization")
+            raise
+        
+        # Log and re-raise other errors
         logger.error(f"Error instantiating toolkit {toolkit_name} with client: {str(e)}")
         raise
 

alita_sdk/runtime/utils/utils.py

@@ -1,6 +1,11 @@
+import base64
+import logging
 import re
 from enum import Enum
+from typing import Any
 
+# DEPRECATED: Tool names no longer use prefixes
+# Kept for backward compatibility only
 TOOLKIT_SPLITTER = "___"
 
 class IndexerKeywords(Enum):
@@ -9,6 +14,12 @@ class IndexerKeywords(Enum):
     PARENT = 'parent_id'
     # DEPENDENCY_ID = 'dependency_id'
     UPDATED_ON = 'updated_on'
+    CONTENT_IN_BYTES = 'loader_content'
+    CONTENT_FILE_NAME = 'loader_content_type'
+    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_.-]')
@@ -18,3 +29,40 @@ def clean_string(s: str) -> str:
     # Replace these characters with an empty string
     cleaned_string = re.sub(clean_string_pattern, '', s)
     return cleaned_string
+
+
+def clean_node_str(s: str) -> str:
+    """Cleans a node string by removing all non-alphanumeric characters except underscores and spaces."""
+    cleaned_string = re.sub(r'[^\w\s]', '', s)
+    return cleaned_string
+
+
+def resolve_image_from_cache(client: Any, cached_image_id: str) -> bytes:
+    """
+    Resolve cached_image_id from client's image cache and return decoded binary data.
+
+    Args:
+        client: AlitaClient instance with _generated_images_cache attribute
+        cached_image_id: The cached image ID to resolve
+
+    Returns:
+        bytes: Decoded binary image data
+
+    Raises:
+        ValueError: If cached_image_id not found or decoding fails
+    """
+    cache = getattr(client, '_generated_images_cache', {})
+
+    if cached_image_id not in cache:
+        raise ValueError(f"Image reference '{cached_image_id}' not found. The image may have expired.")
+
+    cached_data = cache[cached_image_id]
+    base64_data = cached_data.get('base64_data', '')
+    logging.debug(f"Resolved cached_image_id '{cached_image_id}' from cache (length: {len(base64_data)} chars)")
+    # Decode base64 to binary data for image files
+    try:
+        binary_data = base64.b64decode(base64_data)
+        logging.debug(f"Decoded base64 to binary data ({len(binary_data)} bytes)")
+        return binary_data
+    except Exception as e:
+        raise ValueError(f"Failed to decode image data for '{cached_image_id}': {e}")

alita_sdk/tools/__init__.py

@@ -13,6 +13,30 @@ AVAILABLE_TOOLS = {}
 AVAILABLE_TOOLKITS = {}
 FAILED_IMPORTS = {}
 
+
+def _inject_toolkit_id(tool_conf: dict, toolkit_tools) -> None:
+    """Inject `toolkit_id` into tools that expose `api_wrapper.toolkit_id`.
+
+    This reads 'id' from the tool configuration and, if it is an integer,
+    assigns it to the 'toolkit_id' attribute of the 'api_wrapper' for each
+    tool in 'toolkit_tools' that supports it.
+
+    Args:
+        tool_conf: Raw tool configuration item from 'tools_list'.
+        toolkit_tools: List of instantiated tools produced by a toolkit.
+    """
+    toolkit_id = tool_conf.get('id')
+    if isinstance(toolkit_id, int):
+        for t in toolkit_tools:
+            if hasattr(t, 'api_wrapper') and hasattr(t.api_wrapper, 'toolkit_id'):
+                t.api_wrapper.toolkit_id = toolkit_id
+    else:
+        logger.error(
+            f"Toolkit ID is missing or not an integer for tool "
+            f"`{tool_conf.get('type', '')}` with name `{tool_conf.get('name', '')}`"
+        )
+
+
 def _safe_import_tool(tool_name, module_path, get_tools_name=None, toolkit_class_name=None):
     """Safely import a tool module and register available functions/classes."""
     try:
@@ -21,6 +45,12 @@ def _safe_import_tool(tool_name, module_path, get_tools_name=None, toolkit_class
         imported = {}
         if get_tools_name and hasattr(module, get_tools_name):
             imported['get_tools'] = getattr(module, get_tools_name)
+        
+        if hasattr(module, 'get_toolkit'):
+            imported['get_toolkit'] = getattr(module, 'get_toolkit')
+
+        if hasattr(module, 'get_toolkit_available_tools'):
+            imported['get_toolkit_available_tools'] = getattr(module, 'get_toolkit_available_tools')
 
         if toolkit_class_name and hasattr(module, toolkit_class_name):
             imported['toolkit_class'] = getattr(module, toolkit_class_name)
@@ -34,9 +64,10 @@ def _safe_import_tool(tool_name, module_path, get_tools_name=None, toolkit_class
         FAILED_IMPORTS[tool_name] = str(e)
         logger.debug(f"Failed to import {tool_name}: {e}")
 
+
 # Safe imports for all tools
 _safe_import_tool('github', 'github', 'get_tools', 'AlitaGitHubToolkit')
-_safe_import_tool('openapi', 'openapi', 'get_tools')
+_safe_import_tool('openapi', 'openapi', 'get_tools', 'AlitaOpenAPIToolkit')
 _safe_import_tool('jira', 'jira', 'get_tools', 'JiraToolkit')
 _safe_import_tool('confluence', 'confluence', 'get_tools', 'ConfluenceToolkit')
 _safe_import_tool('service_now', 'servicenow', 'get_tools', 'ServiceNowToolkit')
@@ -67,11 +98,11 @@ _safe_import_tool('aws', 'cloud.aws', None, 'AWSToolkit')
 _safe_import_tool('azure', 'cloud.azure', None, 'AzureToolkit')
 _safe_import_tool('gcp', 'cloud.gcp', None, 'GCPToolkit')
 _safe_import_tool('k8s', 'cloud.k8s', None, 'KubernetesToolkit')
-_safe_import_tool('custom_open_api', 'custom_open_api', None, 'OpenApiToolkit')
+# _safe_import_tool('custom_open_api', 'custom_open_api', None, 'OpenApiToolkit')
 _safe_import_tool('elastic', 'elastic', None, 'ElasticToolkit')
 _safe_import_tool('keycloak', 'keycloak', None, 'KeycloakToolkit')
 _safe_import_tool('localgit', 'localgit', None, 'AlitaLocalGitToolkit')
-_safe_import_tool('pandas', 'pandas', 'get_tools', 'PandasToolkit')
+# pandas toolkit removed - use Data Analysis internal tool instead
 _safe_import_tool('azure_search', 'azure_ai.search', 'get_tools', 'AzureSearchToolkit')
 _safe_import_tool('figma', 'figma', 'get_tools', 'FigmaToolkit')
 _safe_import_tool('salesforce', 'salesforce', 'get_tools', 'SalesforceToolkit')
@@ -79,7 +110,6 @@ _safe_import_tool('carrier', 'carrier', 'get_tools', 'AlitaCarrierToolkit')
 _safe_import_tool('ocr', 'ocr', 'get_tools', 'OCRToolkit')
 _safe_import_tool('pptx', 'pptx', 'get_tools', 'PPTXToolkit')
 _safe_import_tool('postman', 'postman', 'get_tools', 'PostmanToolkit')
-_safe_import_tool('memory', 'memory', 'get_tools', 'MemoryToolkit')
 _safe_import_tool('zephyr_squad', 'zephyr_squad', 'get_tools', 'ZephyrSquadToolkit')
 _safe_import_tool('zephyr_essential', 'zephyr_essential', 'get_tools', 'ZephyrEssentialToolkit')
 _safe_import_tool('slack', 'slack', 'get_tools', 'SlackToolkit')
@@ -91,62 +121,81 @@ available_count = len(AVAILABLE_TOOLS)
 total_attempted = len(AVAILABLE_TOOLS) + len(FAILED_IMPORTS)
 logger.info(f"Tool imports completed: {available_count}/{total_attempted} successful")
 
+# Import community module to trigger community toolkit registration
+try:
+    from alita_sdk import community  # noqa: F401
+    logger.debug("Community toolkits registered successfully")
+except ImportError as e:
+    logger.debug(f"Community module not available: {e}")
+
+
 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
-        tool_type = tool['type']
+        toolkit_tools = []
+        settings = tool.get('settings')
 
-        # Handle special cases for ADO tools
-        if tool_type in ['ado_boards', 'ado_wiki', 'ado_plans']:
-            tools.extend(AVAILABLE_TOOLS['ado']['get_tools'](tool_type, tool))
+        # Skip tools without settings early
+        if not settings:
+            logger.warning(f"Tool '{tool.get('type', '')}' has no settings, skipping...")
+            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]:
-            try:
-                get_tools_func = AVAILABLE_TOOLS[tool_type]['get_tools']
-                tools.extend(get_tools_func(tool))
+        # 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 '_'")
 
-            except Exception as e:
-                logger.error(f"Error getting tools for {tool_type}: {e}")
-                raise ToolException(f"Error getting tools for {tool_type}: {e}")
+        # 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'):
+            # Use tool id if available, otherwise use toolkit_name or type as fallback
+            collection_id = tool.get('id') or tool.get('toolkit_name') or tool_type
+            settings['pgvector_configuration']['collection_schema'] = str(collection_id)
 
-        # Handle ADO repos special case (it might be requested as azure_devops_repos)
+        # Handle ADO special cases
+        if tool_type in ['ado_boards', 'ado_wiki', 'ado_plans']:
+            toolkit_tools.extend(AVAILABLE_TOOLS['ado']['get_tools'](tool_type, tool))
         elif tool_type in ['ado_repos', 'azure_devops_repos'] and 'ado_repos' in AVAILABLE_TOOLS:
             try:
-                get_tools_func = AVAILABLE_TOOLS['ado_repos']['get_tools']
-                tools.extend(get_tools_func(tool))
+                toolkit_tools.extend(AVAILABLE_TOOLS['ado_repos']['get_tools'](tool))
             except Exception as e:
                 logger.error(f"Error getting ADO repos tools: {e}")
-
-        # Handle custom modules
-        elif tool.get("settings", {}).get("module"):
+        elif tool_type == 'mcp':
+            logger.debug(f"Skipping MCP toolkit '{tool.get('toolkit_name')}' - handled by runtime toolkit system")
+        elif tool_type == 'planning':
+            logger.debug(f"Skipping planning toolkit '{tool.get('toolkit_name')}' - handled by runtime toolkit system")
+        elif tool_type in AVAILABLE_TOOLS and 'get_tools' in AVAILABLE_TOOLS[tool_type]:
+            try:
+                toolkit_tools.extend(AVAILABLE_TOOLS[tool_type]['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}")
+        elif 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())
+                toolkit_tools.extend(toolkit.get_tools())
             except Exception as e:
                 logger.error(f"Error in getting custom toolkit: {e}")
-
         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}")
+        #
+        # Always inject toolkit_id to each tool 
+        _inject_toolkit_id(tool, toolkit_tools)
+        tools.extend(toolkit_tools)
 
     return tools
 
@@ -166,6 +215,18 @@ def get_toolkits():
     logger.info(f"Successfully loaded {len(toolkit_configs)} toolkit configurations")
     return toolkit_configs
 
+def instantiate_toolkit(tool_config):
+    """Instantiate a toolkit from its configuration."""
+    tool_type = tool_config.get('type')
+    
+    if tool_type in AVAILABLE_TOOLS:
+        tool_module = AVAILABLE_TOOLS[tool_type]
+        
+        if 'get_toolkit' in tool_module:
+             return tool_module['get_toolkit'](tool_config)
+             
+    raise ValueError(f"Toolkit type '{tool_type}' does not support direct instantiation or is not available.")
+
 def get_available_tools():
     """Return list of available tool types."""
     return list(AVAILABLE_TOOLS.keys())
@@ -182,6 +243,42 @@ def get_available_toolkit_models():
     """Return dict with available toolkit classes."""
     return deepcopy(AVAILABLE_TOOLS)
 
+
+def get_toolkit_available_tools(toolkit_type: str, settings: dict) -> dict:
+    """Return dynamic available tools + per-tool JSON schemas for a toolkit instance.
+
+    This is the single SDK entrypoint used by backend services (e.g. indexer_worker)
+    when the UI needs spec/instance-dependent tool enumeration. Toolkits that don't
+    support dynamic enumeration should return an empty payload.
+
+    Args:
+        toolkit_type: toolkit type string (e.g. 'openapi')
+        settings: persisted toolkit settings
+
+    Returns:
+        {
+          "tools": [{"name": str, "description": str}],
+          "args_schemas": {"tool_name": <json schema dict>}
+        }
+    """
+    toolkit_type = (toolkit_type or '').strip().lower()
+    if not isinstance(settings, dict):
+        settings = {}
+
+    tool_module = AVAILABLE_TOOLS.get(toolkit_type) or {}
+    enumerator = tool_module.get('get_toolkit_available_tools')
+    if not callable(enumerator):
+        return {"tools": [], "args_schemas": {}}
+
+    try:
+        result = enumerator(settings)
+        if not isinstance(result, dict):
+            return {"tools": [], "args_schemas": {}, "error": "Invalid response from toolkit enumerator"}
+        return result
+    except Exception as e:  # pylint: disable=W0718
+        logger.exception("Failed to compute available tools for toolkit_type=%s", toolkit_type)
+        return {"tools": [], "args_schemas": {}, "error": str(e)}
+
 def diagnose_imports():
     """Print diagnostic information about tool imports."""
     available_count = len(AVAILABLE_TOOLS)
@@ -218,6 +315,7 @@ def diagnose_imports():
 __all__ = [
     'get_tools',
     'get_toolkits',
+    'get_toolkit_available_tools',
     'get_available_tools',
     'get_failed_imports',
     'get_available_toolkits',

alita_sdk/tools/ado/__init__.py

@@ -9,16 +9,16 @@ def get_tools(tool_type, tool):
     config_dict = {
         # common
         "selected_tools": tool['settings'].get('selected_tools', []),
-        "organization_url": tool['settings']['organization_url'],
         "ado_configuration": tool['settings']['ado_configuration'],
         "limit": tool['settings'].get('limit', 5),
         "toolkit_name": tool.get('toolkit_name', ''),
         # indexer settings
+        "alita": tool['settings'].get('alita', None),
         "llm": tool['settings'].get('llm', None),
         "pgvector_configuration": tool['settings'].get('pgvector_configuration', {}),
         "collection_name": tool['toolkit_name'],
         "doctype": 'doc',
-        "embedding_model": tool['settings'].get('embedding_configuration', {}).get('name', None),
+        "embedding_model": tool['settings'].get('embedding_model'),
         "vectorstore_type": "PGVector"
     }
     if tool_type == 'ado_plans':

alita_sdk/tools/ado/repos/__init__.py

@@ -5,16 +5,18 @@ from pydantic import BaseModel, Field, create_model
 
 import requests
 
+from ...elitea_base import filter_missconfigured_index_tools
 from ....configurations.ado import AdoReposConfiguration
 from ....configurations.pgvector import PgVectorConfiguration
 from ...base.tool import BaseAction
 from .repos_wrapper import ReposApiWrapper
-from ...utils import clean_string, TOOLKIT_SPLITTER, get_max_toolkit_length, check_connection_response
+from ...utils import clean_string, get_max_toolkit_length, check_connection_response
+from ....runtime.utils.constants import TOOLKIT_NAME_META, TOOL_NAME_META, TOOLKIT_TYPE_META
 
 name = "ado_repos"
 
 
-def _get_toolkit(tool) -> BaseToolkit:
+def get_toolkit(tool) -> BaseToolkit:
     return AzureDevOpsReposToolkit().get_toolkit(
         selected_tools=tool['settings'].get('selected_tools', []),
         ado_repos_configuration=tool['settings']['ado_repos_configuration'],
@@ -23,26 +25,21 @@ def _get_toolkit(tool) -> BaseToolkit:
         active_branch=tool['settings'].get('active_branch', ""),
         toolkit_name=tool['settings'].get('toolkit_name', ""),
         pgvector_configuration=tool['settings'].get('pgvector_configuration', {}),
-        collection_name=tool['toolkit_name'],
-        doctype='code',
         embedding_model=tool['settings'].get('embedding_model'),
-        vectorstore_type="PGVector",
+        collection_name=tool['toolkit_name'],
+        alita=tool['settings'].get('alita', None),
+        llm=tool['settings'].get('llm', None),
     )
 
-def get_toolkit():
-    return AzureDevOpsReposToolkit.toolkit_config_schema()
-
 def get_tools(tool):
-    return _get_toolkit(tool).get_tools()
+    return get_toolkit(tool).get_tools()
 
 class AzureDevOpsReposToolkit(BaseToolkit):
     tools: List[BaseTool] = []
-    toolkit_max_length: int = 0
 
     @staticmethod
     def toolkit_config_schema() -> BaseModel:
         selected_tools = {x['name']: x['args_schema'].schema() for x in ReposApiWrapper.model_construct().get_available_tools()}
-        AzureDevOpsReposToolkit.toolkit_max_length = get_max_toolkit_length(selected_tools)
         m = create_model(
             name,
             ado_repos_configuration=(AdoReposConfiguration, Field(description="Ado Repos configuration", default=None,
@@ -62,11 +59,6 @@ class AzureDevOpsReposToolkit(BaseToolkit):
                     "icon_url": "ado-repos-icon.svg",
                     "categories": ["code repositories"],
                     "extra_categories": ["code", "repository", "version control"],
-                    # "configuration_group": {
-                    #     "name": "ado_repos",
-                    #     "label": "Azure DevOps Repositories",
-                    #     "icon_url": "ado-repos-icon.svg",
-                    # }
                 }}}
         )
 
@@ -83,6 +75,7 @@ class AzureDevOpsReposToolkit(BaseToolkit):
         return m
 
     @classmethod
+    @filter_missconfigured_index_tools
     def get_toolkit(cls, selected_tools: list[str] | None = None, toolkit_name: Optional[str] = None, **kwargs):
         from os import environ
 
@@ -101,17 +94,21 @@ class AzureDevOpsReposToolkit(BaseToolkit):
         azure_devops_repos_wrapper = ReposApiWrapper(**wrapper_payload)
         available_tools = azure_devops_repos_wrapper.get_available_tools()
         tools = []
-        prefix = clean_string(toolkit_name, cls.toolkit_max_length) + TOOLKIT_SPLITTER if toolkit_name else ''
         for tool in available_tools:
             if selected_tools:
                 if tool["name"] not in selected_tools:
                     continue
+            description = tool["description"] + f"\nADO instance: {azure_devops_repos_wrapper.organization_url}/{azure_devops_repos_wrapper.project}"
+            if toolkit_name:
+                description = f"{description}\nToolkit: {toolkit_name}"
+            description = description[:1000]
             tools.append(
                 BaseAction(
                     api_wrapper=azure_devops_repos_wrapper,
-                    name=prefix + tool["name"],
-                    description=tool["description"] + f"\nADO instance: {azure_devops_repos_wrapper.organization_url}/{azure_devops_repos_wrapper.project}",
+                    name=tool["name"],
+                    description=description,
                     args_schema=tool["args_schema"],
+                    metadata={TOOLKIT_NAME_META: toolkit_name, TOOLKIT_TYPE_META: name, TOOL_NAME_META: tool["name"]} if toolkit_name else {TOOL_NAME_META: tool["name"]}
                 )
             )
         return cls(tools=tools)

alita_sdk/tools/ado/repos/repos_wrapper.py

@@ -24,7 +24,8 @@ from msrest.authentication import BasicAuthentication
 from pydantic import Field, PrivateAttr, create_model, model_validator, SecretStr
 
 from ..utils import extract_old_new_pairs, generate_diff, get_content_from_generator
-from ...elitea_base import BaseCodeToolApiWrapper
+from ...code_indexer_toolkit import CodeIndexerToolkit
+from ...utils.available_tools_decorator import extend_with_parent_available_tools
 
 logger = logging.getLogger(__name__)
 
@@ -110,8 +111,7 @@ class ArgsSchema(Enum):
             Field(
                 description=(
                     "Branch to be used for read file operation."
-                ),
-                default=None
+                )
             ),
         )
     )
@@ -159,6 +159,7 @@ class ArgsSchema(Enum):
             Field(
                 description="""List of comments, where each comment is a dictionary specifying details about the comment,
                 e.g. [{'file_path': 'src/main.py', 'comment_text': 'Logic needs improvement', 'right_line': 20}]""",
+                default=None,
                 examples=[
                     {
                         "file_path": "src/main.py",
@@ -241,7 +242,7 @@ class ArgsSchema(Enum):
     )
 
 
-class ReposApiWrapper(BaseCodeToolApiWrapper):
+class ReposApiWrapper(CodeIndexerToolkit):
     # TODO use ado_repos_configuration fields
     organization_url: Optional[str]
     project: Optional[str]
@@ -251,15 +252,6 @@ class ReposApiWrapper(BaseCodeToolApiWrapper):
     token: Optional[SecretStr]
     _client: Optional[GitClient] = PrivateAttr()
 
-    llm: Optional[Any] = None
-    # Vector store configuration
-    connection_string: Optional[SecretStr] = None
-    collection_name: Optional[str] = None
-    doctype: Optional[str] = 'code'
-    embedding_model: Optional[str] = "HuggingFaceEmbeddings"
-    embedding_model_params: Optional[dict] = {"model_name": "sentence-transformers/all-MiniLM-L6-v2"}
-    vectorstore_type: Optional[str] = "PGVector"
-
     class Config:
         arbitrary_types_allowed = True
 
@@ -301,7 +293,7 @@ class ReposApiWrapper(BaseCodeToolApiWrapper):
             if not branch_exists(active_branch):
                 raise ToolException(f"The active branch '{active_branch}' does not exist.")
 
-        return values
+        return super().validate_toolkit(values)
 
     def _get_commits(self, file_path: str, branch: str, top: int = None) -> List[GitCommitRef]:
         """
@@ -652,6 +644,9 @@ class ReposApiWrapper(BaseCodeToolApiWrapper):
 
         return dumps(data)
 
+    def download_file(self, path):
+        return b"".join(self._client.get_item_content(self.repository_id, path=path, project=self.project, download=True))
+
     def get_file_content(self, commit_id, path):
         version_descriptor = GitVersionDescriptor(
             version=commit_id, version_type="commit"
@@ -1179,9 +1174,10 @@ class ReposApiWrapper(BaseCodeToolApiWrapper):
         except Exception as e:
             return ToolException(f"Unable to retrieve commits due to error:\n{str(e)}")
 
+    @extend_with_parent_available_tools
     def get_available_tools(self):
         """Return a list of available tools."""
-        tools = [
+        return [
             {
                 "ref": self.list_branches_in_repo,
                 "name": "list_branches_in_repo",
@@ -1272,8 +1268,4 @@ class ReposApiWrapper(BaseCodeToolApiWrapper):
                 "description": self.get_commits.__doc__,
                 "args_schema": ArgsSchema.GetCommits.value,
             },
-        ]        # Add vector search tools from base class (includes index_data + search tools)
-        vector_search_tools = self._get_vector_search_tools()
-        tools.extend(vector_search_tools)
-
-        return tools
+        ]