vectara-agentic 0.1.17__py3-none-any.whl → 0.1.18__py3-none-any.whl

vectara_agentic/__init__.py

@@ -3,7 +3,7 @@ vectara_agentic package.
 """
 
 # Define the package version
-__version__ = "0.1.17"
+__version__ = "0.1.18"
 
 # Import classes and functions from modules
 # from .module1 import Class1, function1

vectara_agentic/_prompts.py

@@ -24,9 +24,11 @@ GENERAL_INSTRUCTIONS = """
 - If you are provided with database tools use them for analytical queries (such as counting, calculating max, min, average, sum, or other statistics).
   For each database, the database tools include: x_list_tables, x_load_data, x_describe_tables, and x_load_sample_data, where 'x' in the database name.
   The x_list_tables tool provides a list of available tables in the x database.
-  Always use the x_describe_tables tool to understand the schema of each table, before you access data from that table.
-  Always use the x_load_sample_data tool to understand the column names, and the unique values in each column, so you can use them in your queries.
-  Some times the user may ask for a specific column value, but the actual value in the table may be different, and you will need to use the correct value.
+  Before issuing a SQL query, always:
+  - Use the x_describe_tables tool to understand the schema of each table
+  - Use the x_load_unique_values tool to understand the unique values in each column.
+    Sometimes the user may ask for a specific column value, but the actual value in the table may be different, and you will need to use the correct value.
+  - Use the x_load_sample_data tool to understand the column names, and typical values in each column.
 - Never call x_load_data to retrieve values from each row in the table.
 - Do not mention table names or database names in your response.
 """
@@ -89,6 +91,8 @@ NEVER surround your response with markdown code markers. You may use code marker
 
 Please use a valid JSON format for the Action Input. Do NOT do this {{'input': 'hello world', 'num_beams': 5}}.
 
+Do not include the Action Input in a wrapper dictionary 'properties' like this: {{'properties': {{'input': 'hello world', 'num_beams': 5}} }}.
+
 If this format is used, the user will respond in the following format:
 
 ```

vectara_agentic/db_tools.py

@@ -0,0 +1,74 @@
+"""
+This module contains the code to extend and improve DatabaseToolSpec
+Specifically adding load_sample_data and load_unique_values methods, as well as
+making sure the load_data method returns a list of text values from the database, not Document[] objects.
+"""
+from abc import ABC
+from typing import Callable, Any
+
+#
+# Additional database tool
+#
+class DBTool(ABC):
+    """
+    A base class for vectara-agentic database tools extensions
+    """
+    def __init__(self, load_data_fn: Callable):
+        self.load_data_fn = load_data_fn
+
+class DBLoadData(DBTool):
+    """
+    A tool to Run SQL query on the database and return the result.
+    """
+    def __call__(self, query: str) -> Any:
+        """Query and load data from the Database, returning a list of Documents.
+
+        Args:
+            query (str): an SQL query to filter tables and rows.
+
+        Returns:
+            List[text]: a list of text values from the database.
+        """
+        res = self.load_data_fn(query)
+        return [d.text for d in res]
+
+class DBLoadSampleData(DBTool):
+    """
+    A tool to load a sample of data from the specified database table.
+
+    This tool fetches the first num_rows (default 25) rows from the given table
+    using a provided database query function.
+    """
+    def __call__(self, table_name: str, num_rows: int = 25) -> Any:
+        """
+        Fetches the first num_rows rows from the specified database table.
+
+        Args:
+            table_name (str): The name of the database table.
+
+        Returns:
+            Any: The result of the database query.
+        """
+        return self.load_data_fn(f"SELECT * FROM {table_name} LIMIT {num_rows}")
+
+class DBLoadUniqueValues(DBTool):
+    """
+    A tool to list all unique values for each column in a set of columns of a database table.
+    """
+    def __call__(self, table_name: str, columns: list[str], num_vals: int = 200) -> dict:
+        """
+        Fetches the first num_vals unique values from the specified columns of the database table.
+
+        Args:
+            table_name (str): The name of the database table.
+            columns (list[str]): The list of columns to fetch unique values from.
+            num_vals (int): The number of unique values to fetch for each column. Default is 200.
+
+        Returns:
+            dict: A dictionary containing the unique values for each column.
+        """
+        res = {}
+        for column in columns:
+            unique_vals = self.load_data_fn(f'SELECT DISTINCT "{column}" FROM {table_name} LIMIT {num_vals}')
+            res[column] = [d.text for d in unique_vals]
+        return res

vectara_agentic/tools.py

@@ -7,7 +7,7 @@ import re
 import importlib
 import os
 
-from typing import Callable, List, Any, Optional, Type
+from typing import Callable, List, Dict, Any, Optional, Type
 from pydantic import BaseModel, Field
 
 from llama_index.core.tools import FunctionTool
@@ -18,7 +18,8 @@ from llama_index.core.tools.types import ToolMetadata, ToolOutput
 
 
 from .types import ToolType
-from .tools_catalog import summarize_text, rephrase_text, critique_text, get_bad_topics, DBLoadSampleData
+from .tools_catalog import summarize_text, rephrase_text, critique_text, get_bad_topics
+from .db_tools import DBLoadSampleData, DBLoadUniqueValues, DBLoadData
 
 LI_packages = {
     "yahoo_finance": ToolType.QUERY,
@@ -42,9 +43,15 @@ LI_packages = {
         },
         "GoogleSearchToolSpec": {"google_search": ToolType.QUERY},
     },
+    "slack": {
+        "SlackToolSpec": {
+            "load_data": ToolType.QUERY,
+            "send_message": ToolType.ACTION,
+            "fetch_channel": ToolType.QUERY,
+        }
+    }
 }
 
-
 class VectaraToolMetadata(ToolMetadata):
     """
     A subclass of ToolMetadata adding the tool_type attribute.
@@ -152,6 +159,8 @@ class VectaraToolFactory:
         reranker: str = "mmr",
         rerank_k: int = 50,
         mmr_diversity_bias: float = 0.2,
+        udf_expression: str = None,
+        rerank_chain: List[Dict] = None,
         include_citations: bool = True,
         fcs_threshold: float = 0.0,
     ) -> VectaraTool:
@@ -171,10 +180,16 @@ class VectaraToolFactory:
             reranker (str, optional): The reranker mode.
             rerank_k (int, optional): Number of top-k documents for reranking.
             mmr_diversity_bias (float, optional): MMR diversity bias.
+            udf_expression (str, optional): the user defined expression for reranking results.
+            rerank_chain (List[Dict], optional): A list of rerankers to be applied sequentially.
+                Each dictionary should specify the "type" of reranker (mmr, slingshot, udf)
+                and any other parameters (e.g. "limit" or "cutoff" for any type,
+                "diversity_bias" for mmr, and "user_function" for udf).
+            If using slingshot/multilingual_reranker_v1, it must be first in the list.
             include_citations (bool, optional): Whether to include citations in the response.
                 If True, uses markdown vectara citations that requires the Vectara scale plan.
             fcs_threshold (float, optional): a threshold for factual consistency.
-                If set above 0, the tool notifies the calling agent that it "cannot respond" if FCS is too low
+                If set above 0, the tool notifies the calling agent that it "cannot respond" if FCS is too low.
 
         Returns:
             VectaraTool: A VectaraTool object.
@@ -218,6 +233,8 @@ class VectaraToolFactory:
                 reranker=reranker,
                 rerank_k=rerank_k if rerank_k * self.num_corpora <= 100 else int(100 / self.num_corpora),
                 mmr_diversity_bias=mmr_diversity_bias,
+                udf_expression=udf_expression,
+                rerank_chain=rerank_chain,
                 n_sentence_before=n_sentences_before,
                 n_sentence_after=n_sentences_after,
                 lambda_val=lambda_val,
@@ -495,10 +512,22 @@ class ToolsFactory:
                     tool.metadata.description + f"The database tables include data about {content_description}."
                 )
 
-        # Update load_data_tool to return only text instead of "Document" objects (to save on space)
+        # Add two new tools: load_sample_data and load_unique_values
         load_data_tool_index = next(i for i, t in enumerate(tools) if t.metadata.name.endswith("load_data"))
-        sample_data_fn = DBLoadSampleData(tools[load_data_tool_index])
+        load_data_fn_original = tools[load_data_tool_index].fn
+
+        load_data_fn = DBLoadData(load_data_fn_original)
+        load_data_fn.__name__ = f"{tool_name_prefix}_load_data"
+        load_data_tool = self.create_tool(load_data_fn, ToolType.QUERY)
+
+        sample_data_fn = DBLoadSampleData(load_data_fn_original)
         sample_data_fn.__name__ = f"{tool_name_prefix}_load_sample_data"
         sample_data_tool = self.create_tool(sample_data_fn, ToolType.QUERY)
-        tools.append(sample_data_tool)
+
+        load_unique_values_fn = DBLoadUniqueValues(load_data_fn_original)
+        load_unique_values_fn.__name__ = f"{tool_name_prefix}_load_unique_values"
+        load_unique_values_tool = self.create_tool(load_unique_values_fn, ToolType.QUERY)
+
+        tools[load_data_tool_index] = load_data_tool
+        tools.extend([sample_data_tool, load_unique_values_tool])
         return tools

vectara_agentic/tools_catalog.py

@@ -1,8 +1,7 @@
 """
 This module contains the tools catalog for the Vectara Agentic.
 """
-
-from typing import Callable, Any, List
+from typing import List
 from functools import lru_cache
 from pydantic import Field
 import requests
@@ -114,7 +113,7 @@ def critique_text(
 
 
 #
-# Guardrails tools
+# Guardrails tool: returns list of topics to avoid
 #
 def get_bad_topics() -> List[str]:
     """
@@ -128,36 +127,3 @@ def get_bad_topics() -> List[str]:
         "adult content",
         "illegal activities",
     ]
-
-
-#
-# Additional database tool
-#
-class DBLoadSampleData:
-    """
-    A tool to load a sample of data from the specified database table.
-
-    This tool fetches the first num_rows (default 25) rows from the given table
-    using a provided database query function.
-    """
-
-    def __init__(self, load_data_tool: Callable):
-        """
-        Initializes the DBLoadSampleData object with the provided load_data_tool function.
-
-        Args:
-            load_data_tool (Callable): A function to execute the SQL query.
-        """
-        self.load_data_tool = load_data_tool
-
-    def __call__(self, table_name: str, num_rows: int = 25) -> Any:
-        """
-        Fetches the first num_rows rows from the specified database table.
-
-        Args:
-            table_name (str): The name of the database table.
-
-        Returns:
-            Any: The result of the database query.
-        """
-        return self.load_data_tool(f"SELECT * FROM {table_name} LIMIT {num_rows}")

{vectara_agentic-0.1.17.dist-info → vectara_agentic-0.1.18.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: vectara_agentic
-Version: 0.1.17
+Version: 0.1.18
 Summary: A Python package for creating AI Assistants and AI Agents with Vectara
 Home-page: https://github.com/vectara/py-vectara-agentic
 Author: Ofer Mendelevitch
@@ -17,7 +17,7 @@ Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: llama-index==0.11.20
-Requires-Dist: llama-index-indices-managed-vectara==0.2.2
+Requires-Dist: llama-index-indices-managed-vectara==0.2.3
 Requires-Dist: llama-index-agent-llm-compiler==0.2.0
 Requires-Dist: llama-index-agent-openai==0.3.4
 Requires-Dist: llama-index-llms-openai==0.2.16
@@ -33,6 +33,7 @@ Requires-Dist: llama-index-tools-database==0.2.0
 Requires-Dist: llama-index-tools-google==0.2.0
 Requires-Dist: llama-index-tools-tavily-research==0.2.0
 Requires-Dist: llama-index-tools-neo4j==0.2.0
+Requires-Dist: llama-index-tools-slack==0.2.0
 Requires-Dist: tavily-python==0.5.0
 Requires-Dist: yahoo-finance==1.4.0
 Requires-Dist: openinference-instrumentation-llama-index==3.0.2
@@ -182,12 +183,21 @@ print(response)
 - `stock_news`: provides news about a company
 - `stock_analyst_recommendations`: provides stock analyst recommendations for a company.
 
-1. **Database tools**: providing tools to inspect and query a database
+4. **Database tools**: providing tools to inspect and query a database
 - `list_tables`: list all tables in the database
 - `describe_tables`: describe the schema of tables in the database
 - `load_data`: returns data based on a SQL query
+- `load_sample_data`: returns the first 25 rows of a table
+- `load_unique_values`: returns the top unique values for a given column
 
-More tools coming soon...
+In addition, we include various other tools from LlamaIndex ToolSpecs:
+* Tavily search
+* arxiv
+* neo4j
+* Google tools (including gmail, calendar, and search)
+* Slack
+
+Note that some of these tools may require API keys as environment variables
 
 You can create your own tool directly from a Python function using the `create_tool()` method of the `ToolsFactory` class:
 

{vectara_agentic-0.1.17.dist-info → vectara_agentic-0.1.18.dist-info}/RECORD

@@ -1,18 +1,19 @@
 tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 tests/test_agent.py,sha256=aQYYr_8hKlFiDgyI5Dd39TG5vkmDJe7F_nIzMTCLsTQ,2517
 tests/test_tools.py,sha256=hDAlXkWKuXHnAjeQwMuTLTwNdRsM-xR7muBzFkZRefw,2942
-vectara_agentic/__init__.py,sha256=FzVPx4Ti8Q5wVyUpNpsVuuRLjoyHkR563ncsfextd2k,508
+vectara_agentic/__init__.py,sha256=anqWxWDvo9azuZJN7qgh8CgTS13vCXtOtetc0wV7WOc,508
 vectara_agentic/_callback.py,sha256=EexD7-Qx2lZuQk4kjzwvIJAyfIzroWKz2VaVPD4TTkM,4621
 vectara_agentic/_observability.py,sha256=v0xxTk8KI8nVK2rpyGqOVhyva9ymqOmZK5brKqFOwMM,3828
-vectara_agentic/_prompts.py,sha256=CxSKueS4lF130VMSovFKeapTM9DZ8Ja4akEFIw7q0Ks,5670
+vectara_agentic/_prompts.py,sha256=EMSiTriaKQLiD0gPiGFAriOII2cfr2_b-_8qkD7UUz0,5848
 vectara_agentic/agent.py,sha256=9nzPVb16wpeT5RiWQJiO5LuR9VTYp8yScKUqqgem7h8,19963
 vectara_agentic/agent_endpoint.py,sha256=I3zTEezbAiNeW5I41r0NjIaR8Ucn4oe1XVcALekakaA,1959
-vectara_agentic/tools.py,sha256=_rXEK9tuuDMM_Zf04AoehklydTMSb1esqEYkNjLvmyY,20103
-vectara_agentic/tools_catalog.py,sha256=CpFTTyocsTu-Sx8KjqSsZZhLditSrV-vRZd2YMFU0AY,4940
+vectara_agentic/db_tools.py,sha256=QFMU1A08n-ql4eVygfp-RtSKrPATq3JtPUTIr5nZZIQ,2604
+vectara_agentic/tools.py,sha256=HLAiBglGEkrOBXmUK4EKi60hNDmzhHGwK9855HxcViQ,21590
+vectara_agentic/tools_catalog.py,sha256=5NlJypdu0IKa7mODxVOwo05lw3PqQJtSl_ZOsUDH_TA,3986
 vectara_agentic/types.py,sha256=FbZXc5oPje6kdimfrksDc8F-tYHSLK8ReAv7O291YkI,1131
 vectara_agentic/utils.py,sha256=BCd5tI9LKVXuPc4WsxGwgbouLcWr200JDWSsGNUwZg0,3822
-vectara_agentic-0.1.17.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
-vectara_agentic-0.1.17.dist-info/METADATA,sha256=iFCMBcWNtYPnyR2RhBAnVvu-8uQWNzqJx8N3tSKL05k,13595
-vectara_agentic-0.1.17.dist-info/WHEEL,sha256=OVMc5UfuAQiSplgO0_WdW7vXVGAt9Hdd6qtN4HotdyA,91
-vectara_agentic-0.1.17.dist-info/top_level.txt,sha256=Y7TQTFdOYGYodQRltUGRieZKIYuzeZj2kHqAUpfCUfg,22
-vectara_agentic-0.1.17.dist-info/RECORD,,
+vectara_agentic-0.1.18.dist-info/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
+vectara_agentic-0.1.18.dist-info/METADATA,sha256=1lsOej02Xt3sMXFgLO1JGpEXY26U6GUoSxjR0bWKYFo,13990
+vectara_agentic-0.1.18.dist-info/WHEEL,sha256=P9jw-gEje8ByB7_hXoICnHtVCrEwMQh-630tKvQWehc,91
+vectara_agentic-0.1.18.dist-info/top_level.txt,sha256=Y7TQTFdOYGYodQRltUGRieZKIYuzeZj2kHqAUpfCUfg,22
+vectara_agentic-0.1.18.dist-info/RECORD,,

{vectara_agentic-0.1.17.dist-info → vectara_agentic-0.1.18.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (75.2.0)
+Generator: setuptools (75.3.0)
 Root-Is-Purelib: true
 Tag: py3-none-any