alita-sdk 0.3.435__py3-none-any.whl → 0.3.457__py3-none-any.whl

alita_sdk/runtime/utils/mcp_sse_client.py

@@ -0,0 +1,405 @@
+"""
+MCP SSE (Server-Sent Events) Client
+Handles persistent SSE connections for MCP servers like Atlassian
+"""
+import asyncio
+import json
+import logging
+from typing import Dict, Any, Optional, AsyncIterator
+import aiohttp
+
+logger = logging.getLogger(__name__)
+
+
+class McpSseClient:
+    """
+    Client for MCP servers using SSE (Server-Sent Events) transport.
+    
+    For Atlassian-style SSE (dual-connection model):
+    - GET request opens persistent SSE stream for receiving events
+    - POST requests send commands (return 202 Accepted immediately)
+    - Responses come via the GET stream
+    
+    This client handles:
+    - Opening persistent SSE connection via GET
+    - Sending JSON-RPC requests via POST
+    - Reading SSE event streams
+    - Matching responses to requests by ID
+    """
+    
+    def __init__(self, url: str, session_id: str, headers: Optional[Dict[str, str]] = None, timeout: int = 300):
+        """
+        Initialize SSE client.
+        
+        Args:
+            url: Base URL of the MCP SSE server
+            session_id: Client-generated UUID for session
+            headers: Additional headers (e.g., Authorization)
+            timeout: Request timeout in seconds
+        """
+        self.url = url
+        self.session_id = session_id
+        self.headers = headers or {}
+        self.timeout = timeout
+        self.url_with_session = f"{url}?sessionId={session_id}"
+        self._stream_task = None
+        self._pending_requests = {}  # request_id -> asyncio.Future
+        self._stream_session = None
+        self._stream_response = None
+        self._endpoint_ready = asyncio.Event()  # Signal when endpoint is received
+        
+        logger.info(f"[MCP SSE Client] Initialized for {url} with session {session_id}")
+    
+    async def _ensure_stream_connected(self):
+        """Ensure the GET stream is connected and reading events."""
+        if self._stream_task is None or self._stream_task.done():
+            logger.info(f"[MCP SSE Client] Opening persistent SSE stream...")
+            self._stream_session = aiohttp.ClientSession(timeout=aiohttp.ClientTimeout(total=None))
+            
+            headers = {
+                "Accept": "text/event-stream",
+                **self.headers
+            }
+            
+            self._stream_response = await self._stream_session.get(self.url_with_session, headers=headers)
+            
+            logger.info(f"[MCP SSE Client] Stream opened: status={self._stream_response.status}")
+            
+            # Handle 401 Unauthorized - need OAuth
+            if self._stream_response.status == 401:
+                from ..utils.mcp_oauth import (
+                    McpAuthorizationRequired,
+                    canonical_resource,
+                    extract_resource_metadata_url,
+                    fetch_resource_metadata_async,
+                    infer_authorization_servers_from_realm,
+                    fetch_oauth_authorization_server_metadata
+                )
+                
+                auth_header = self._stream_response.headers.get('WWW-Authenticate', '')
+                resource_metadata_url = extract_resource_metadata_url(auth_header, self.url)
+                
+                metadata = None
+                if resource_metadata_url:
+                    metadata = await fetch_resource_metadata_async(
+                        resource_metadata_url,
+                        session=self._stream_session,
+                        timeout=30
+                    )
+                
+                # Infer authorization servers if not in metadata
+                if not metadata or not metadata.get('authorization_servers'):
+                    inferred_servers = infer_authorization_servers_from_realm(auth_header, self.url)
+                    if inferred_servers:
+                        if not metadata:
+                            metadata = {}
+                        metadata['authorization_servers'] = inferred_servers
+                        logger.info(f"[MCP SSE Client] Inferred authorization servers: {inferred_servers}")
+                        
+                        # Fetch OAuth metadata
+                        auth_server_metadata = fetch_oauth_authorization_server_metadata(inferred_servers[0], timeout=30)
+                        if auth_server_metadata:
+                            metadata['oauth_authorization_server'] = auth_server_metadata
+                            logger.info(f"[MCP SSE Client] Fetched OAuth metadata")
+                
+                raise McpAuthorizationRequired(
+                    message=f"MCP server {self.url} requires OAuth authorization",
+                    server_url=canonical_resource(self.url),
+                    resource_metadata_url=resource_metadata_url,
+                    www_authenticate=auth_header,
+                    resource_metadata=metadata,
+                    status=self._stream_response.status,
+                    tool_name=self.url,
+                )
+            
+            if self._stream_response.status != 200:
+                error_text = await self._stream_response.text()
+                raise Exception(f"Failed to open SSE stream: HTTP {self._stream_response.status}: {error_text}")
+            
+            # Start background task to read stream
+            self._stream_task = asyncio.create_task(self._read_stream())
+    
+    async def _read_stream(self):
+        """Background task that continuously reads the SSE stream."""
+        logger.info(f"[MCP SSE Client] Starting stream reader...")
+        
+        try:
+            buffer = ""
+            current_event = {}
+            
+            async for chunk in self._stream_response.content.iter_chunked(1024):
+                chunk_str = chunk.decode('utf-8')
+                buffer += chunk_str
+                
+                # Process complete lines
+                while '\n' in buffer:
+                    line, buffer = buffer.split('\n', 1)
+                    line_str = line.strip()
+                    
+                    # Empty line indicates end of event
+                    if not line_str:
+                        if current_event and 'data' in current_event:
+                            self._process_event(current_event)
+                            current_event = {}
+                        continue
+                    
+                    # Parse SSE fields
+                    if line_str.startswith('event:'):
+                        current_event['event'] = line_str[6:].strip()
+                    elif line_str.startswith('data:'):
+                        data_str = line_str[5:].strip()
+                        current_event['data'] = data_str
+                    elif line_str.startswith('id:'):
+                        current_event['id'] = line_str[3:].strip()
+                    
+        except Exception as e:
+            logger.error(f"[MCP SSE Client] Stream reader error: {e}")
+            # Fail all pending requests
+            for future in self._pending_requests.values():
+                if not future.done():
+                    future.set_exception(e)
+        finally:
+            logger.info(f"[MCP SSE Client] Stream reader stopped")
+    
+    def _process_event(self, event: Dict[str, str]):
+        """Process a complete SSE event."""
+        event_type = event.get('event', 'message')
+        data_str = event.get('data', '')
+        
+        # Handle 'endpoint' event - server provides the actual session URL to use
+        if event_type == 'endpoint':
+            # Extract session ID from endpoint URL
+            # Format: /v1/sse?sessionId=<uuid>
+            if 'sessionId=' in data_str:
+                new_session_id = data_str.split('sessionId=')[1].split('&')[0]
+                logger.info(f"[MCP SSE Client] Server provided session ID: {new_session_id}")
+                self.session_id = new_session_id
+                self.url_with_session = f"{self.url}?sessionId={new_session_id}"
+                self._endpoint_ready.set()  # Signal that we can now send requests
+            return
+        
+        # Skip other non-message events
+        if event_type != 'message' and not data_str.startswith('{'):
+            return
+        
+        if not data_str:
+            return
+        
+        try:
+            data = json.loads(data_str)
+            request_id = data.get('id')
+            
+            logger.debug(f"[MCP SSE Client] Received response for request {request_id}")
+            
+            # Resolve pending request
+            if request_id and request_id in self._pending_requests:
+                future = self._pending_requests.pop(request_id)
+                if not future.done():
+                    future.set_result(data)
+                
+        except json.JSONDecodeError as e:
+            logger.warning(f"[MCP SSE Client] Failed to parse SSE data: {e}, data: {repr(data_str)[:200]}")
+                            
+        except Exception as e:
+            logger.error(f"[MCP SSE Client] Stream reader error: {e}")
+            # Fail all pending requests
+            for future in self._pending_requests.values():
+                if not future.done():
+                    future.set_exception(e)
+        finally:
+            logger.info(f"[MCP SSE Client] Stream reader stopped")
+    
+    async def send_request(self, method: str, params: Optional[Dict[str, Any]] = None, request_id: Optional[str] = None) -> Dict[str, Any]:
+        """
+        Send a JSON-RPC request and wait for response via SSE stream.
+        
+        Uses dual-connection model:
+        1. GET stream is kept open to receive responses
+        2. POST request sends the command (returns 202 immediately)
+        3. Response comes via the GET stream
+        
+        Args:
+            method: JSON-RPC method name (e.g., "tools/list", "tools/call")
+            params: Method parameters
+            request_id: Optional request ID (auto-generated if not provided)
+            
+        Returns:
+            Parsed JSON-RPC response
+            
+        Raises:
+            Exception: If request fails or times out
+        """
+        import time
+        if request_id is None:
+            request_id = f"{method.replace('/', '_')}_{int(time.time() * 1000)}"
+        
+        request = {
+            "jsonrpc": "2.0",
+            "id": request_id,
+            "method": method,
+            "params": params or {}
+        }
+        
+        logger.debug(f"[MCP SSE Client] Sending request: {method} (id={request_id})")
+        
+        # Ensure stream is connected
+        await self._ensure_stream_connected()
+        
+        # Wait for endpoint event (server provides the actual session ID to use)
+        await asyncio.wait_for(self._endpoint_ready.wait(), timeout=10)
+        
+        # Create future for this request
+        future = asyncio.Future()
+        self._pending_requests[request_id] = future
+        
+        # Send POST request
+        headers = {
+            "Content-Type": "application/json",
+            **self.headers
+        }
+        
+        timeout = aiohttp.ClientTimeout(total=30)
+        
+        try:
+            async with aiohttp.ClientSession(timeout=timeout) as session:
+                async with session.post(self.url_with_session, json=request, headers=headers) as response:
+                    if response.status == 404:
+                        error_text = await response.text()
+                        raise Exception(f"HTTP 404: {error_text}")
+                    
+                    # 202 is expected - response will come via stream
+                    if response.status not in [200, 202]:
+                        error_text = await response.text()
+                        raise Exception(f"HTTP {response.status}: {error_text}")
+            
+            # Wait for response from stream (with timeout)
+            result = await asyncio.wait_for(future, timeout=self.timeout)
+            
+            # Check for JSON-RPC error
+            if 'error' in result:
+                error = result['error']
+                raise Exception(f"MCP Error: {error.get('message', str(error))}")
+            
+            return result
+                    
+        except asyncio.TimeoutError:
+            self._pending_requests.pop(request_id, None)
+            logger.error(f"[MCP SSE Client] Request timeout after {self.timeout}s")
+            raise Exception(f"SSE request timeout after {self.timeout}s")
+        except Exception as e:
+            self._pending_requests.pop(request_id, None)
+            logger.error(f"[MCP SSE Client] Request failed: {e}")
+            raise
+    
+    async def close(self):
+        """Close the persistent SSE stream."""
+        logger.info(f"[MCP SSE Client] Closing connection...")
+        
+        # Cancel background stream reader task
+        if self._stream_task and not self._stream_task.done():
+            self._stream_task.cancel()
+            try:
+                await self._stream_task
+            except (asyncio.CancelledError, Exception) as e:
+                logger.debug(f"[MCP SSE Client] Stream task cleanup: {e}")
+        
+        # Close response stream
+        if self._stream_response and not self._stream_response.closed:
+            try:
+                self._stream_response.close()
+            except Exception as e:
+                logger.debug(f"[MCP SSE Client] Response close error: {e}")
+        
+        # Close session
+        if self._stream_session and not self._stream_session.closed:
+            try:
+                await self._stream_session.close()
+                # Give aiohttp time to cleanup
+                await asyncio.sleep(0.1)
+            except Exception as e:
+                logger.debug(f"[MCP SSE Client] Session close error: {e}")
+        
+        logger.info(f"[MCP SSE Client] Connection closed")
+    
+    async def __aenter__(self):
+        """Async context manager entry."""
+        return self
+    
+    async def __aexit__(self, exc_type, exc_val, exc_tb):
+        """Async context manager exit."""
+        await self.close()
+    
+    async def initialize(self) -> Dict[str, Any]:
+        """
+        Send initialize request to establish MCP protocol session.
+        
+        Returns:
+            Server capabilities and info
+        """
+        response = await self.send_request(
+            method="initialize",
+            params={
+                "protocolVersion": "2024-11-05",
+                "capabilities": {
+                    "roots": {"listChanged": True},
+                    "sampling": {}
+                },
+                "clientInfo": {
+                    "name": "Alita MCP Client",
+                    "version": "1.0.0"
+                }
+            }
+        )
+        
+        logger.info(f"[MCP SSE Client] MCP session initialized")
+        return response.get('result', {})
+    
+    async def list_tools(self) -> list:
+        """
+        Discover available tools from the MCP server.
+        
+        Returns:
+            List of tool definitions
+        """
+        response = await self.send_request(method="tools/list")
+        result = response.get('result', {})
+        tools = result.get('tools', [])
+        
+        logger.info(f"[MCP SSE Client] Discovered {len(tools)} tools")
+        return tools
+    
+    async def list_prompts(self) -> list:
+        """
+        Discover available prompts from the MCP server.
+        
+        Returns:
+            List of prompt definitions
+        """
+        response = await self.send_request(method="prompts/list")
+        result = response.get('result', {})
+        prompts = result.get('prompts', [])
+        
+        logger.debug(f"[MCP SSE Client] Discovered {len(prompts)} prompts")
+        return prompts
+    
+    async def call_tool(self, tool_name: str, arguments: Dict[str, Any]) -> Any:
+        """
+        Execute a tool on the MCP server.
+        
+        Args:
+            tool_name: Name of the tool to call
+            arguments: Tool arguments
+            
+        Returns:
+            Tool execution result
+        """
+        response = await self.send_request(
+            method="tools/call",
+            params={
+                "name": tool_name,
+                "arguments": arguments
+            }
+        )
+        
+        result = response.get('result', {})
+        return result

alita_sdk/runtime/utils/toolkit_utils.py

@@ -29,13 +29,14 @@ def instantiate_toolkit_with_client(toolkit_config: Dict[str, Any],
         
     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:
@@ -70,6 +71,12 @@ 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/tools/ado/repos/__init__.py

@@ -63,6 +63,7 @@ class AzureDevOpsReposToolkit(BaseToolkit):
                     "icon_url": "ado-repos-icon.svg",
                     "categories": ["code repositories"],
                     "extra_categories": ["code", "repository", "version control"],
+                    "max_length": AzureDevOpsReposToolkit.toolkit_max_length
                 }}}
         )
 

alita_sdk/tools/ado/test_plan/__init__.py

@@ -27,7 +27,6 @@ class AzureDevOpsPlansToolkit(BaseToolkit):
         AzureDevOpsPlansToolkit.toolkit_max_length = get_max_toolkit_length(selected_tools)
         m = create_model(
             name_alias,
-            name=(str, Field(description="Toolkit name", json_schema_extra={'toolkit_name': True, 'max_toolkit_length': AzureDevOpsPlansToolkit.toolkit_max_length})),
             ado_configuration=(AdoConfiguration, Field(description="Ado configuration", json_schema_extra={'configuration_types': ['ado']})),
             limit=(Optional[int], Field(description="ADO plans limit used for limitation of the list with results", default=5)),
             # indexer settings
@@ -40,6 +39,7 @@ class AzureDevOpsPlansToolkit(BaseToolkit):
                 {
                     "label": "ADO plans",
                     "icon_url": "ado-plans.svg",
+                    "max_length": AzureDevOpsPlansToolkit.toolkit_max_length,
                     "categories": ["test management"],
                     "extra_categories": ["test case management", "qa"],
                     "sections": {

alita_sdk/tools/ado/wiki/__init__.py

@@ -24,11 +24,6 @@ class AzureDevOpsWikiToolkit(BaseToolkit):
         AzureDevOpsWikiToolkit.toolkit_max_length = get_max_toolkit_length(selected_tools)
         m = create_model(
             name_alias,
-            name=(str, Field(description="Toolkit name",
-                             json_schema_extra={
-                                 'toolkit_name': True,
-                                 'max_toolkit_length': AzureDevOpsWikiToolkit.toolkit_max_length})
-                  ),
             ado_configuration=(AdoConfiguration, Field(description="Ado configuration", json_schema_extra={'configuration_types': ['ado']})),
             # indexer settings
             pgvector_configuration=(Optional[PgVectorConfiguration], Field(default=None,
@@ -42,6 +37,7 @@ class AzureDevOpsWikiToolkit(BaseToolkit):
                     'metadata': {
                         "label": "ADO wiki",
                         "icon_url": "ado-wiki-icon.svg",
+                        "max_length": AzureDevOpsWikiToolkit.toolkit_max_length,
                         "categories": ["documentation"],
                         "extra_categories": ["knowledge base", "documentation management", "wiki"],
                         "sections": {

alita_sdk/tools/ado/work_item/__init__.py

@@ -23,11 +23,6 @@ class AzureDevOpsWorkItemsToolkit(BaseToolkit):
         AzureDevOpsWorkItemsToolkit.toolkit_max_length = get_max_toolkit_length(selected_tools)
         m = create_model(
             name,
-            name=(str, Field(description="Toolkit name",
-                             json_schema_extra={
-                                 'toolkit_name': True,
-                                 'max_toolkit_length': AzureDevOpsWorkItemsToolkit.toolkit_max_length})
-                  ),
             ado_configuration=(AdoConfiguration, Field(description="Ado Work Item configuration", json_schema_extra={'configuration_types': ['ado']})),
             limit=(Optional[int], Field(description="ADO plans limit used for limitation of the list with results", default=5)),
             selected_tools=(List[Literal[tuple(selected_tools)]], Field(default=[], json_schema_extra={'args_schemas': selected_tools})),
@@ -42,6 +37,7 @@ class AzureDevOpsWorkItemsToolkit(BaseToolkit):
                     'metadata': {
                         "label": "ADO boards",
                         "icon_url": "ado-boards-icon.svg",
+                        "max_length": AzureDevOpsWorkItemsToolkit.toolkit_max_length,
                         "categories": ["project management"],
                         "extra_categories": ["work item management", "issue tracking", "agile boards"],
                         "sections": {

alita_sdk/tools/base_indexer_toolkit.py

@@ -15,6 +15,8 @@ from ..runtime.utils.utils import IndexerKeywords
 
 logger = logging.getLogger(__name__)
 
+DEFAULT_CUT_OFF = 0.2
+
 # Base Vector Store Schema Models
 BaseIndexParams = create_model(
     "BaseIndexParams",
@@ -37,7 +39,7 @@ BaseSearchParams = create_model(
         default={},
         examples=["{\"key\": \"value\"}", "{\"status\": \"active\"}"]
     )),
-    cut_off=(Optional[float], Field(description="Cut-off score for search results", default=0.5, ge=0, le=1)),
+    cut_off=(Optional[float], Field(description="Cut-off score for search results", default=DEFAULT_CUT_OFF, ge=0, le=1)),
     search_top=(Optional[int], Field(description="Number of top results to return", default=10)),
     full_text_search=(Optional[Dict[str, Any]], Field(
         description="Full text search parameters. Can be a dictionary with search options.",
@@ -67,7 +69,7 @@ BaseStepbackSearchParams = create_model(
         default={},
         examples=["{\"key\": \"value\"}", "{\"status\": \"active\"}"]
     )),
-    cut_off=(Optional[float], Field(description="Cut-off score for search results", default=0.5, ge=0, le=1)),
+    cut_off=(Optional[float], Field(description="Cut-off score for search results", default=DEFAULT_CUT_OFF, ge=0, le=1)),
     search_top=(Optional[int], Field(description="Number of top results to return", default=10)),
     full_text_search=(Optional[Dict[str, Any]], Field(
         description="Full text search parameters. Can be a dictionary with search options.",
@@ -380,7 +382,7 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
     def search_index(self,
                      query: str,
                      index_name: str = "",
-                     filter: dict | str = {}, cut_off: float = 0.5,
+                     filter: dict | str = {}, cut_off: float = DEFAULT_CUT_OFF,
                      search_top: int = 10, reranker: dict = {},
                      full_text_search: Optional[Dict[str, Any]] = None,
                      reranking_config: Optional[Dict[str, Dict[str, Any]]] = None,
@@ -411,7 +413,7 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
                      query: str,
                      messages: List[Dict[str, Any]] = [],
                      index_name: str = "",
-                     filter: dict | str = {}, cut_off: float = 0.5,
+                     filter: dict | str = {}, cut_off: float = DEFAULT_CUT_OFF,
                      search_top: int = 10, reranker: dict = {},
                      full_text_search: Optional[Dict[str, Any]] = None,
                      reranking_config: Optional[Dict[str, Dict[str, Any]]] = None,
@@ -436,7 +438,7 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
                      query: str,
                      messages: List[Dict[str, Any]] = [],
                      index_name: str = "",
-                     filter: dict | str = {}, cut_off: float = 0.5,
+                     filter: dict | str = {}, cut_off: float = DEFAULT_CUT_OFF,
                      search_top: int = 10, reranker: dict = {},
                      full_text_search: Optional[Dict[str, Any]] = None,
                      reranking_config: Optional[Dict[str, Dict[str, Any]]] = None,
@@ -470,6 +472,7 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
                 "collection": index_name,
                 "type": IndexerKeywords.INDEX_META_TYPE.value,
                 "indexed": 0,
+                "updated": 0,
                 "state": IndexerKeywords.INDEX_META_IN_PROGRESS.value,
                 "index_configuration": index_configuration,
                 "created_on": created_on,
@@ -487,7 +490,8 @@ class BaseIndexerToolkit(VectorStoreWrapperBase):
         #
         if index_meta_raw:
             metadata = copy.deepcopy(index_meta_raw.get("metadata", {}))
-            metadata["indexed"] = result
+            metadata["indexed"] = self.get_indexed_count(index_name)
+            metadata["updated"] = result
             metadata["state"] = state
             metadata["updated_on"] = time.time()
             #

alita_sdk/tools/bitbucket/__init__.py

@@ -61,6 +61,7 @@ class AlitaBitbucketToolkit(BaseToolkit):
                 'metadata':
                     {
                         "label": "Bitbucket", "icon_url": "bitbucket-icon.svg",
+                        "max_length": AlitaBitbucketToolkit.toolkit_max_length,
                         "categories": ["code repositories"],
                         "extra_categories": ["bitbucket", "git", "repository", "code", "version control"],
                     }

alita_sdk/tools/code/sonar/__init__.py

@@ -29,7 +29,7 @@ class SonarToolkit(BaseToolkit):
         SonarToolkit.toolkit_max_length = get_max_toolkit_length(selected_tools)
         return create_model(
             name,
-            sonar_project_name=(str, Field(description="Project name of the desired repository", json_schema_extra={'toolkit_name': True, 'max_toolkit_length': SonarToolkit.toolkit_max_length})),
+            sonar_project_name=(str, Field(description="Project name of the desired repository")),
             sonar_configuration=(SonarConfiguration, Field(description="Sonar Configuration", json_schema_extra={'configuration_types': ['sonar']})),
             selected_tools=(List[Literal[tuple(selected_tools)]], Field(default=[], json_schema_extra={'args_schemas': selected_tools})),
             __config__=ConfigDict(json_schema_extra=

alita_sdk/tools/confluence/__init__.py

@@ -67,8 +67,7 @@ class ConfluenceToolkit(BaseToolkit):
 
         model = create_model(
             name,
-            space=(str, Field(description="Space", json_schema_extra={'toolkit_name': True,
-                                                                                    'max_toolkit_length': ConfluenceToolkit.toolkit_max_length})),
+            space=(str, Field(description="Space")),
             cloud=(bool, Field(description="Hosting Option", json_schema_extra={'configuration': True})),
             limit=(int, Field(description="Pages limit per request", default=5)),
             labels=(Optional[str], Field(
@@ -95,6 +94,7 @@ class ConfluenceToolkit(BaseToolkit):
                 'metadata': {
                     "label": "Confluence",
                     "icon_url": None,
+                    "max_length": ConfluenceToolkit.toolkit_max_length,
                     "categories": ["documentation"],
                     "extra_categories": ["confluence", "wiki", "knowledge base", "documentation", "atlassian"]
                 }

alita_sdk/tools/github/__init__.py

@@ -53,6 +53,7 @@ class AlitaGitHubToolkit(BaseToolkit):
                     'metadata': {
                         "label": "GitHub",
                         "icon_url": None,
+                        "max_length": AlitaGitHubToolkit.toolkit_max_length,
                         "categories": ["code repositories"],
                         "extra_categories": ["github", "git", "repository", "code", "version control"],
                     },
@@ -62,8 +63,7 @@ class AlitaGitHubToolkit(BaseToolkit):
                                                              json_schema_extra={'configuration_types': ['github']})),
             pgvector_configuration=(Optional[PgVectorConfiguration], Field(description="PgVector configuration", default=None,
                                                                      json_schema_extra={'configuration_types': ['pgvector']})),
-            repository=(str, Field(description="Github repository", json_schema_extra={'toolkit_name': True,
-                                                                                       'max_toolkit_length': AlitaGitHubToolkit.toolkit_max_length})),
+            repository=(str, Field(description="Github repository")),
             active_branch=(Optional[str], Field(description="Active branch", default="main")),
             base_branch=(Optional[str], Field(description="Github Base branch", default="main")),
             # embedder settings

alita_sdk/tools/gitlab/__init__.py

@@ -43,7 +43,7 @@ class AlitaGitlabToolkit(BaseToolkit):
         AlitaGitlabToolkit.toolkit_max_length = get_max_toolkit_length(selected_tools)
         return create_model(
             name,
-            repository=(str, Field(description="GitLab repository", json_schema_extra={'toolkit_name': True, 'max_toolkit_length': AlitaGitlabToolkit.toolkit_max_length})),
+            repository=(str, Field(description="GitLab repository")),
             gitlab_configuration=(GitlabConfiguration, Field(description="GitLab configuration", json_schema_extra={'configuration_types': ['gitlab']})),
             branch=(str, Field(description="Main branch", default="main")),
             # indexer settings
@@ -57,6 +57,7 @@ class AlitaGitlabToolkit(BaseToolkit):
                 'metadata': {
                     "label": "GitLab",
                     "icon_url": None,
+                    "max_length": AlitaGitlabToolkit.toolkit_max_length,
                     "categories": ["code repositories"],
                     "extra_categories": ["gitlab", "git", "repository", "code", "version control"],
                 }

alita_sdk/tools/gitlab_org/__init__.py

@@ -30,8 +30,6 @@ class AlitaGitlabSpaceToolkit(BaseToolkit):
         AlitaGitlabSpaceToolkit.toolkit_max_length = get_max_toolkit_length(selected_tools)
         return create_model(
             name,
-            name=(str, Field(description="Toolkit name", json_schema_extra={'toolkit_name': True,
-                                                                            'max_toolkit_length': AlitaGitlabSpaceToolkit.toolkit_max_length})),
             gitlab_configuration=(GitlabConfiguration, Field(description="GitLab configuration",
                                                                        json_schema_extra={
                                                                            'configuration_types': ['gitlab']})),
@@ -46,6 +44,7 @@ class AlitaGitlabSpaceToolkit(BaseToolkit):
                 'metadata': {
                     "label": "GitLab Org",
                     "icon_url": None,
+                    "max_length": AlitaGitlabSpaceToolkit.toolkit_max_length,
                     "categories": ["code repositories"],
                     "extra_categories": ["gitlab", "git", "repository", "code", "version control"],
                 }