camel-ai 0.2.72a6__py3-none-any.whl → 0.2.72a8__py3-none-any.whl

camel/__init__.py

@@ -14,7 +14,7 @@
 
 from camel.logger import disable_logging, enable_logging, set_log_level
 
-__version__ = '0.2.72a6'
+__version__ = '0.2.72a8'
 
 __all__ = [
     '__version__',

camel/agents/chat_agent.py

@@ -375,6 +375,11 @@ class ChatAgent(BaseAgent):
         pause_event (Optional[asyncio.Event]): Event to signal pause of the
             agent's operation. When clear, the agent will pause its execution.
             (default: :obj:`None`)
+        prune_tool_calls_from_memory (bool): Whether to clean tool
+            call messages from memory after response generation to save token
+            usage. When enabled, removes FUNCTION/TOOL role messages and
+            ASSISTANT messages with tool_calls after each step.
+            (default: :obj:`False`)
     """
 
     def __init__(
@@ -411,6 +416,7 @@ class ChatAgent(BaseAgent):
         tool_execution_timeout: Optional[float] = None,
         mask_tool_output: bool = False,
         pause_event: Optional[asyncio.Event] = None,
+        prune_tool_calls_from_memory: bool = False,
     ) -> None:
         if isinstance(model, ModelManager):
             self.model_backend = model
@@ -492,6 +498,7 @@ class ChatAgent(BaseAgent):
         self._image_retry_count: Dict[str, int] = {}
         # Store images to attach to next user message
         self.pause_event = pause_event
+        self.prune_tool_calls_from_memory = prune_tool_calls_from_memory
 
     def reset(self):
         r"""Resets the :obj:`ChatAgent` to its initial state."""
@@ -1472,6 +1479,10 @@ class ChatAgent(BaseAgent):
 
         self._record_final_output(response.output_messages)
 
+        # Clean tool call messages from memory after response generation
+        if self.prune_tool_calls_from_memory and tool_call_records:
+            self.memory.clean_tool_calls()
+
         return self._convert_to_chatagent_response(
             response,
             tool_call_records,
@@ -1730,6 +1741,10 @@ class ChatAgent(BaseAgent):
 
         self._record_final_output(response.output_messages)
 
+        # Clean tool call messages from memory after response generation
+        if self.prune_tool_calls_from_memory and tool_call_records:
+            self.memory.clean_tool_calls()
+
         return self._convert_to_chatagent_response(
             response,
             tool_call_records,
@@ -3270,11 +3285,20 @@ class ChatAgent(BaseAgent):
             return
 
         # Start async streaming response
+        last_response = None
         async for response in self._astream_response(
             openai_messages, num_tokens, response_format
         ):
+            last_response = response
             yield response
 
+        # Clean tool call messages from memory after response generation
+        if self.prune_tool_calls_from_memory and last_response:
+            # Extract tool_calls from the last response info
+            tool_calls = last_response.info.get("tool_calls", [])
+            if tool_calls:
+                self.memory.clean_tool_calls()
+
     async def _astream_response(
         self,
         openai_messages: List[OpenAIMessage],
@@ -3799,6 +3823,7 @@ class ChatAgent(BaseAgent):
             stop_event=self.stop_event,
             tool_execution_timeout=self.tool_execution_timeout,
             pause_event=self.pause_event,
+            prune_tool_calls_from_memory=self.prune_tool_calls_from_memory,
         )
 
         # Copy memory if requested

camel/memories/agent_memories.py

@@ -88,6 +88,44 @@ class ChatHistoryMemory(AgentMemory):
     def clear(self) -> None:
         self._chat_history_block.clear()
 
+    def clean_tool_calls(self) -> None:
+        r"""Removes tool call messages from memory.
+        This method removes all FUNCTION/TOOL role messages and any ASSISTANT
+        messages that contain tool_calls in their meta_dict to save token
+        usage.
+        """
+        from camel.types import OpenAIBackendRole
+
+        # Get all messages from storage
+        record_dicts = self._chat_history_block.storage.load()
+        if not record_dicts:
+            return
+
+        # Filter out tool-related messages
+        cleaned_records = []
+        for record in record_dicts:
+            role = record.get('role_at_backend')
+
+            # Skip FUNCTION messages
+            if role == OpenAIBackendRole.FUNCTION.value:
+                continue
+
+            # Skip TOOL messages
+            if role == OpenAIBackendRole.TOOL.value:
+                continue
+
+            # Skip ASSISTANT messages with tool_calls
+            if role == OpenAIBackendRole.ASSISTANT.value:
+                meta_dict = record.get('meta_dict', {})
+                if meta_dict and 'tool_calls' in meta_dict:
+                    continue
+
+            # Keep all other messages
+            cleaned_records.append(record)
+
+        # Save the cleaned records back to storage
+        self._chat_history_block.storage.save(cleaned_records)
+
 
 class VectorDBMemory(AgentMemory):
     r"""An agent memory wrapper of :obj:`VectorDBBlock`. This memory queries

camel/memories/base.py

@@ -149,6 +149,14 @@ class AgentMemory(MemoryBlock, ABC):
         """
         return self.get_context_creator().create_context(self.retrieve())
 
+    def clean_tool_calls(self) -> None:
+        r"""Removes tool call messages from memory.
+        This is an optional method that can be overridden by subclasses
+        to implement cleaning of tool-related messages. By default, it
+        does nothing, maintaining backward compatibility.
+        """
+        pass
+
     def __repr__(self) -> str:
         r"""Returns a string representation of the AgentMemory.
 

camel/storages/vectordb_storages/__init__.py

@@ -42,4 +42,5 @@ __all__ = [
     'VectorRecord',
     'VectorDBStatus',
     'PgVectorStorage',
+    'SurrealStorage',
 ]

camel/storages/vectordb_storages/surreal.py

@@ -0,0 +1,415 @@
+# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
+# 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.
+# ========= Copyright 2023-2024 @ CAMEL-AI.org. All Rights Reserved. =========
+import re
+from typing import TYPE_CHECKING, Any, Dict, List, Optional
+
+from camel.logger import get_logger
+from camel.storages.vectordb_storages import (
+    BaseVectorStorage,
+    VectorDBQuery,
+    VectorDBQueryResult,
+    VectorDBStatus,
+    VectorRecord,
+)
+from camel.types import VectorDistance
+from camel.utils import dependencies_required
+
+if TYPE_CHECKING:
+    from surrealdb import Surreal  # type: ignore[import-not-found]
+
+logger = get_logger(__name__)
+
+
+class SurrealStorage(BaseVectorStorage):
+    r"""An implementation of the `BaseVectorStorage` using SurrealDB,
+    a scalable, distributed database with WebSocket support, for
+    efficient vector storage and similarity search.
+
+    SurrealDB official site and documentation can be found at:
+    `SurrealDB <https://surrealdb.com>`_
+
+    Args:
+        url (str): WebSocket URL for connecting to SurrealDB
+            (default: "ws://localhost:8000/rpc").
+        table (str): Name of the table used for storing vectors
+            (default: "vector_store").
+        vector_dim (int): Dimensionality of the stored vectors.
+        distance (VectorDistance): Distance metric used for similarity
+            comparisons (default: VectorDistance.COSINE).
+        namespace (str): SurrealDB namespace to use (default: "default").
+        database (str): SurrealDB database name (default: "demo").
+        user (str): Username for authentication (default: "root").
+        password (str): Password for authentication (default: "root").
+
+    Notes:
+        - SurrealDB supports flexible schema and powerful querying capabilities
+        via SQL-like syntax over WebSocket.
+        - This implementation manages connection setup and ensures the target
+        table exists.
+        - Suitable for applications requiring distributed vector storage and
+        search with real-time updates.
+    """
+
+    @dependencies_required('surrealdb')
+    def __init__(
+        self,
+        *,
+        url: str = "ws://localhost:8000/rpc",
+        table: str = "vector_store",
+        vector_dim: int = 786,
+        distance: VectorDistance = VectorDistance.COSINE,
+        namespace: str = "default",
+        database: str = "demo",
+        user: str = "root",
+        password: str = "root",
+    ) -> None:
+        r"""Initialize SurrealStorage with connection settings and ensure
+        the target table exists.
+
+        Args:
+            url (str): WebSocket URL for connecting to SurrealDB.
+                (default: :obj:`"ws://localhost:8000/rpc"`)
+            table (str): Name of the table used for vector storage.
+                (default: :obj:`"vector_store"`)
+            vector_dim (int): Dimensionality of the stored vectors.
+                (default: :obj:`786`)
+            distance (VectorDistance): Distance metric for similarity
+                searches. (default: :obj:`VectorDistance.COSINE`)
+            namespace (str): SurrealDB namespace to use.
+                (default: :obj:`"default"`)
+            database (str): SurrealDB database name.
+                (default: :obj:`"demo"`)
+            user (str): Username for authentication.
+                (default: :obj:`"root"`)
+            password (str): Password for authentication.
+                (default: :obj:`"root"`)
+        """
+
+        self.url = url
+        self.table = table
+        self.ns = namespace
+        self.db = database
+        self.user = user
+        self.password = password
+        self.vector_dim = vector_dim
+        self.distance = distance
+        self._check_and_create_table()
+        self._surreal_client = None
+
+    def _table_exists(self) -> bool:
+        r"""Check whether the target table exists in the database.
+
+        Returns:
+            bool: True if the table exists, False otherwise.
+        """
+        from surrealdb import Surreal  # type: ignore[import-not-found]
+
+        with Surreal(self.url) as db:
+            db.signin({"username": self.user, "password": self.password})
+            db.use(self.ns, self.db)
+            res = db.query_raw("INFO FOR DB;")
+            tables = res['result'][0]['result'].get('tables', {})
+            return self.table in tables
+
+    def _get_table_info(self) -> Dict[str, int]:
+        r"""Retrieve dimension and record count from the table metadata.
+
+        Returns:
+            Dict[str, int]: A dictionary with 'dim' and 'count' keys.
+        """
+        from surrealdb import Surreal  # type: ignore[import-not-found]
+
+        if not self._table_exists():
+            return {"dim": self.vector_dim, "count": 0}
+        with Surreal(self.url) as db:
+            db.signin({"username": self.user, "password": self.password})
+            db.use(self.ns, self.db)
+            res = db.query_raw(f"INFO FOR TABLE {self.table};")
+
+            indexes = res['result'][0]['result'].get("indexes", {})
+
+            dim = self.vector_dim
+            idx_def = indexes.get("hnsw_idx")
+            if idx_def and isinstance(idx_def, str):
+                m = re.search(r"DIMENSION\s+(\d+)", idx_def)
+                if m:
+                    dim = int(m.group(1))
+            cnt = db.query_raw(f"SELECT COUNT() AS count FROM {self.table};")
+            try:
+                count = cnt['result'][0]['result'][0]['count']
+            except (KeyError, IndexError, TypeError):
+                logger.warning(
+                    "Unexpected result format when counting records: %s", cnt
+                )
+                count = 0
+
+            return {"dim": dim, "count": count}
+
+    def _create_table(self):
+        r"""Define and create the vector storage table with HNSW index."""
+        from surrealdb import Surreal  # type: ignore[import-not-found]
+
+        with Surreal(self.url) as db:
+            db.signin({"username": self.user, "password": self.password})
+            db.use(self.ns, self.db)
+            db.query_raw(
+                f"""DEFINE TABLE {self.table} SCHEMALESS;
+                DEFINE FIELD payload    ON {self.table} TYPE object;
+                DEFINE FIELD embedding  ON {self.table} TYPE array;
+                DEFINE INDEX hnsw_idx   ON {self.table}
+                            FIELDS embedding HNSW DIMENSION {self.vector_dim};
+                """
+            )
+        logger.info(f"Table '{self.table}' created successfully.")
+
+    def _drop_table(self):
+        r"""Drop the vector storage table if it exists."""
+        from surrealdb import Surreal  # type: ignore[import-not-found]
+
+        with Surreal(self.url) as db:
+            db.signin({"username": self.user, "password": self.password})
+            db.use(self.ns, self.db)
+            db.query_raw(f"REMOVE TABLE IF EXISTS {self.table};")
+        logger.info(f"Table '{self.table}' deleted successfully.")
+
+    def _check_and_create_table(self):
+        r"""Check if the table exists and matches the expected vector
+        dimension. If not, create a new table.
+        """
+        if self._table_exists():
+            in_dim = self._get_table_info()["dim"]
+            if in_dim != self.vector_dim:
+                raise ValueError(
+                    f"Table {self.table} exists with dimension {in_dim}, "
+                    f"expected {self.vector_dim}"
+                )
+        else:
+            self._create_table()
+
+    def _validate_and_convert_records(
+        self, records: List[VectorRecord]
+    ) -> List[Dict]:
+        r"""Validate and convert VectorRecord instances into
+        SurrealDB-compatible dictionaries.
+
+        Args:
+            records (List[VectorRecord]): List of vector records to insert.
+
+        Returns:
+            List[Dict]: Transformed list of dicts ready for insertion.
+        """
+        validate_data = []
+        for record in records:
+            if len(record.vector) != self.vector_dim:
+                raise ValueError(
+                    f"Vector dimension mismatch: expected {self.vector_dim}, "
+                    f"got {len(record.vector)}"
+                )
+            record_dict = {
+                "payload": record.payload if record.payload else {},
+                "embedding": record.vector,
+            }
+            validate_data.append(record_dict)
+
+        return validate_data
+
+    def query(
+        self,
+        query: VectorDBQuery,
+        **kwargs: Any,
+    ) -> List[VectorDBQueryResult]:
+        r"""Perform a top-k similarity search using the configured distance
+        metric.
+
+        Args:
+            query (VectorDBQuery): Query containing the query vector
+                and top_k value.
+
+        Returns:
+            List[VectorDBQueryResult]: Ranked list of matching records
+                with similarity scores.
+        """
+        from surrealdb import Surreal  # type: ignore[import-not-found]
+
+        metric = {
+            VectorDistance.COSINE: "cosine",
+            VectorDistance.EUCLIDEAN: "euclidean",
+            VectorDistance.DOT: "dot",
+        }[self.distance]
+
+        metric_func = {
+            VectorDistance.COSINE: "vector::similarity::cosine",
+            VectorDistance.EUCLIDEAN: "vector::distance::euclidean",
+            VectorDistance.DOT: "vector::dot",
+        }.get(self.distance)
+
+        if not metric_func:
+            raise ValueError(f"Unsupported distance metric: {self.distance}")
+
+        with Surreal(self.url) as db:
+            db.signin({"username": self.user, "password": self.password})
+            db.use(self.ns, self.db)
+
+            # Use parameterized query to prevent SQL injection
+            sql_query = f"""SELECT payload, embedding,
+                    {metric_func}(embedding, $query_vec) AS score
+                FROM {self.table}
+                WHERE embedding <|{query.top_k},{metric}|> $query_vec
+                ORDER BY score;
+            """
+
+            response = db.query_raw(
+                sql_query, {"query_vec": query.query_vector}
+            )
+
+            if not response.get("result") or not response["result"]:
+                return []
+
+            results = response["result"][0]
+
+            if "result" not in results:
+                return []
+
+            return [
+                VectorDBQueryResult(
+                    record=VectorRecord(
+                        vector=row["embedding"], payload=row.get("payload", {})
+                    ),
+                    similarity=(
+                        1.0 - row["score"]
+                        if self.distance == VectorDistance.COSINE
+                        else 1.0 / (1.0 + row["score"])
+                        if self.distance == VectorDistance.EUCLIDEAN
+                        else row["score"]
+                    ),
+                )
+                for row in results["result"]
+            ]
+
+    def add(self, records: List[VectorRecord], **kwargs) -> None:
+        r"""Insert validated vector records into the SurrealDB table.
+
+        Args:
+            records (List[VectorRecord]): List of vector records to add.
+        """
+        logger.info(
+            "Adding %d records to table '%s'.", len(records), self.table
+        )
+        from surrealdb import Surreal  # type: ignore[import-not-found]
+
+        try:
+            with Surreal(self.url) as db:
+                db.signin({"username": self.user, "password": self.password})
+                db.use(self.ns, self.db)
+
+                validated_records = self._validate_and_convert_records(records)
+                for record in validated_records:
+                    db.create(self.table, record)
+
+            logger.info(
+                "Successfully added %d records to table '%s'.",
+                len(records),
+                self.table,
+            )
+        except Exception as e:
+            logger.error(
+                "Failed to add records to table '%s': %s",
+                self.table,
+                str(e),
+                exc_info=True,
+            )
+            raise
+
+    def delete(
+        self, ids: Optional[List[str]] = None, if_all: bool = False, **kwargs
+    ) -> None:
+        r"""Delete specific records by ID or clear the entire table.
+
+        Args:
+            ids (Optional[List[str]]): List of record IDs to delete.
+            if_all (bool): Whether to delete all records in the table.
+        """
+        from surrealdb import Surreal  # type: ignore[import-not-found]
+        from surrealdb.data.types.record_id import (  # type: ignore[import-not-found]
+            RecordID,
+        )
+
+        try:
+            with Surreal(self.url) as db:
+                db.signin({"username": self.user, "password": self.password})
+                db.use(self.ns, self.db)
+
+                if if_all:
+                    db.delete(self.table, **kwargs)
+                    logger.info(
+                        f"Deleted all records from table '{self.table}'"
+                    )
+                    return
+
+                if not ids:
+                    raise ValueError(
+                        "Either `ids` must be provided or `if_all=True`"
+                    )
+
+                for id_str in ids:
+                    rec = RecordID(self.table, id_str)
+                    db.delete(rec, **kwargs)
+                    logger.info(f"Deleted record {rec}")
+
+        except Exception as e:
+            logger.exception("Error deleting records from SurrealDB")
+            raise RuntimeError(f"Failed to delete records {ids!r}") from e
+
+    def status(self) -> VectorDBStatus:
+        r"""Retrieve the status of the vector table including dimension and
+        count.
+
+        Returns:
+            VectorDBStatus: Object containing vector table metadata.
+        """
+        status = self._get_table_info()
+
+        dim = status.get("dim")
+        count = status.get("count")
+
+        if dim is None or count is None:
+            raise ValueError("Vector dimension and count cannot be None")
+
+        return VectorDBStatus(
+            vector_dim=dim,
+            vector_count=count,
+        )
+
+    def clear(self) -> None:
+        r"""Reset the vector table by dropping and recreating it."""
+        self._drop_table()
+        self._create_table()
+
+    def load(self) -> None:
+        r"""Load the collection hosted on cloud service."""
+        # SurrealDB doesn't require explicit loading
+        raise NotImplementedError("SurrealDB does not support loading")
+
+    @property
+    def client(self) -> "Surreal":
+        r"""Provides access to the underlying SurrealDB client."""
+        if self._surreal_client is None:
+            from surrealdb import Surreal  # type: ignore[import-not-found]
+
+            self._surreal_client = Surreal(self.url)
+            self._surreal_client.signin(  # type: ignore[attr-defined]
+                {"username": self.user, "password": self.password}
+            )
+            self._surreal_client.use(self.ns, self.db)  # type: ignore[attr-defined]
+        return self._surreal_client

camel/toolkits/__init__.py

@@ -85,6 +85,8 @@ from .craw4ai_toolkit import Crawl4AIToolkit
 from .markitdown_toolkit import MarkItDownToolkit
 from .note_taking_toolkit import NoteTakingToolkit
 from .message_agent_toolkit import AgentCommunicationToolkit
+from .web_deploy_toolkit import WebDeployToolkit
+from .screenshot_toolkit import ScreenshotToolkit
 
 __all__ = [
     'BaseToolkit',
@@ -158,4 +160,6 @@ __all__ = [
     'MarkItDownToolkit',
     'NoteTakingToolkit',
     'AgentCommunicationToolkit',
+    'WebDeployToolkit',
+    'ScreenshotToolkit',
 ]

camel/toolkits/human_toolkit.py

@@ -52,7 +52,7 @@ class HumanToolkit(BaseToolkit):
         logger.info(f"User reply: {reply}")
         return reply
 
-    def send_message_to_user(self, message: str) -> None:
+    def send_message_to_user(self, message: str) -> str:
         r"""Use this tool to send a tidy message to the user in one short
         sentence.
 
@@ -68,9 +68,13 @@ class HumanToolkit(BaseToolkit):
 
         Args:
             message (str): The tidy and informative message for the user.
+
+        Returns:
+            str: Confirmation that the message was successfully sent.
         """
         print(f"\nAgent Message:\n{message}")
         logger.info(f"\nAgent Message:\n{message}")
+        return f"Message successfully sent to user: '{message}'"
 
     def get_tools(self) -> List[FunctionTool]:
         r"""Returns a list of FunctionTool objects representing the

camel/toolkits/markitdown_toolkit.py

@@ -33,7 +33,7 @@ class MarkItDownToolkit(BaseToolkit):
     ):
         super().__init__(timeout=timeout)
 
-    def load_files(self, file_paths: List[str]) -> Dict[str, str]:
+    def read_files(self, file_paths: List[str]) -> Dict[str, str]:
         r"""Scrapes content from a list of files and converts it to Markdown.
 
         This function takes a list of local file paths, attempts to convert
@@ -74,5 +74,5 @@ class MarkItDownToolkit(BaseToolkit):
                 representing the functions in the toolkit.
         """
         return [
-            FunctionTool(self.load_files),
+            FunctionTool(self.read_files),
         ]