1
from langchain.agents import AgentExecutor
2
from langchain_cohere.react_multi_hop.agent import create_cohere_react_agent
3
from langchain_core.prompts import ChatPromptTemplate
4
from langchain_cohere.chat_models import ChatCohere
5
from langchain_community.utilities.sql_database import SQLDatabase
6
from langchain_community.agent_toolkits import SQLDatabaseToolkit
7
import os
8
import json

1
# Uncomment if you need to install the following packages
2
#!pip install --quiet langchain langchain_cohere langchain_experimental --upgrade

1
# load the cohere api key
2
os.environ["COHERE_API_KEY"] = ""

1
DB_NAME='Chinook.db'
2
MODEL="command-r-plus"
3
llm = ChatCohere(model=MODEL, temperature=0.1,verbose=True)
4
db = SQLDatabase.from_uri(f"sqlite:///{DB_NAME}")
5
toolkit = SQLDatabaseToolkit(db=db, llm=llm)
6
context = toolkit.get_context()
7
tools = toolkit.get_tools()
8
9
print('**List of pre-defined Langchain Tools**')
10
print([tool.name for tool in tools])

**List of pre-defined Langchain Tools**
['sql_db_query', 'sql_db_schema', 'sql_db_list_tables', 'sql_db_query_checker']

1
# define the prompt template
2
prompt = ChatPromptTemplate.from_template("{input}")
3
# instantiate the ReAct agent
4
agent = create_cohere_react_agent(
5
   llm=llm,
6
   tools=tools,
7
   prompt=prompt,
8
)
9
agent_executor = AgentExecutor(agent=agent,
10
                               tools=tools,
11
                               verbose=True,
12
                               return_intermediate_steps=True
13
                    )

1
output=agent_executor.invoke({
2
   "input": 'what tables are available?',
3
})

> Entering new AgentExecutor chain...

I will use the sql_db_list_tables tool to find out which tables are available.
{'tool_name': 'sql_db_list_tables', 'parameters': {'tool_input': ''}}
Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, TrackRelevant Documents: 0
Cited Documents: 0
Answer: The following tables are available: Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track.
Grounded answer: The following tables are available: <co: 0="">Album</co:>, <co: 0="">Artist</co:>, <co: 0="">Customer</co:>, <co: 0="">Employee</co:>, <co: 0="">Genre</co:>, <co: 0="">Invoice</co:>, <co: 0="">InvoiceLine</co:>, <co: 0="">MediaType</co:>, <co: 0="">Playlist</co:>, <co: 0="">PlaylistTrack</co:>, <co: 0="">Track</co:>.
> Finished chain.

1
print(output['output'])

The following tables are available: Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track.

1
output=agent_executor.invoke({
2
   "input": 'show the first row of the Playlist and Genre tables?',
3
})

> Entering new AgentExecutor chain...

I will use the sql_db_schema tool to find the first row of the Playlist and Genre tables.
{'tool_name': 'sql_db_schema', 'parameters': {'table_names': 'Playlist, Genre'}}

CREATE TABLE "Genre" (
    "GenreId" INTEGER NOT NULL,
    "Name" NVARCHAR(120),
    PRIMARY KEY ("GenreId")
)
/*
3 rows from Genre table:
GenreId	Name
1	Rock
2	Jazz
3	Metal
*/
CREATE TABLE "Playlist" (
    "PlaylistId" INTEGER NOT NULL,
    "Name" NVARCHAR(120),
    PRIMARY KEY ("PlaylistId")
)
/*
3 rows from Playlist table:
PlaylistId	Name
1	Music
2	Movies
3	TV Shows
*/Relevant Documents: 0
Cited Documents: 0
Answer: Here is the first row of the Genre table:
| GenreId | Name |
|---|---|
| 1 | Rock |
Here is the first row of the Playlist table:
| PlaylistId | Name |
|---|---|
| 1 | Music |
Grounded answer: Here is the first row of the Genre table:
| <co: 0="">GenreId</co:> | <co: 0="">Name</co:> |
|---|---|
| <co: 0="">1</co:> | <co: 0="">Rock</co:> |
Here is the first row of the Playlist table:
| <co: 0="">PlaylistId</co:> | <co: 0="">Name</co:> |
|---|---|
| <co: 0="">1</co:> | <co: 0="">Music</co:> |
> Finished chain.

1
print(output['output'])

Here is the first row of the Genre table:
| GenreId | Name |
|---|---|
| 1 | Rock |
Here is the first row of the Playlist table:
| PlaylistId | Name |
|---|---|
| 1 | Music |

1
output=agent_executor.invoke({
2
   "input": 'which countries have the most invoices?',
3
})

> Entering new AgentExecutor chain...

I will search for the number of invoices per country and then write an answer.
{'tool_name': 'sql_db_query', 'parameters': {'query': 'SELECT country, COUNT(*) AS invoice_count FROM invoices GROUP BY country ORDER BY invoice_count DESC'}}
Error: (sqlite3.OperationalError) no such table: invoices
[SQL: SELECT country, COUNT(*) AS invoice_count FROM invoices GROUP BY country ORDER BY invoice_count DESC]
(Background on this error at: https://sqlalche.me/e/20/e3q8)
I received an error message. I will now use the sql_db_list_tables tool to find out the correct table name.
{'tool_name': 'sql_db_list_tables', 'parameters': {'tool_input': ''}}
Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track
I found out that the correct table name is 'Invoice'. I will now rewrite my query and run it again.
{'tool_name': 'sql_db_query', 'parameters': {'query': 'SELECT country, COUNT(*) AS invoice_count FROM Invoice GROUP BY country ORDER BY invoice_count DESC'}}
Error: (sqlite3.OperationalError) no such column: country
[SQL: SELECT country, COUNT(*) AS invoice_count FROM Invoice GROUP BY country ORDER BY invoice_count DESC]
(Background on this error at: https://sqlalche.me/e/20/e3q8)
I received another error message. I will now use the sql_db_schema tool to find out the correct column name.
{'tool_name': 'sql_db_schema', 'parameters': {'table_names': 'Invoice'}}

CREATE TABLE "Invoice" (
    "InvoiceId" INTEGER NOT NULL,
    "CustomerId" INTEGER NOT NULL,
    "InvoiceDate" DATETIME NOT NULL,
    "BillingAddress" NVARCHAR(70),
    "BillingCity" NVARCHAR(40),
    "BillingState" NVARCHAR(40),
    "BillingCountry" NVARCHAR(40),
    "BillingPostalCode" NVARCHAR(10),
    "Total" NUMERIC(10, 2) NOT NULL,
    PRIMARY KEY ("InvoiceId"),
    FOREIGN KEY("CustomerId") REFERENCES "Customer" ("CustomerId")
)
/*
3 rows from Invoice table:
InvoiceId	CustomerId	InvoiceDate	BillingAddress	BillingCity	BillingState	BillingCountry	BillingPostalCode	Total
1	2	2021-01-01 00:00:00	Theodor-Heuss-Straße 34	Stuttgart	None	Germany	70174	1.98
2	4	2021-01-02 00:00:00	Ullevålsveien 14	Oslo	None	Norway	0171	3.96
3	8	2021-01-03 00:00:00	Grétrystraat 63	Brussels	None	Belgium	1000	5.94
*/
I found out that the correct column name is 'BillingCountry'. I will now rewrite my query and run it again.
{'tool_name': 'sql_db_query', 'parameters': {'query': 'SELECT BillingCountry AS country, COUNT(*) AS invoice_count FROM Invoice GROUP BY BillingCountry ORDER BY invoice_count DESC'}}
[('USA', 91), ('Canada', 56), ('France', 35), ('Brazil', 35), ('Germany', 28), ('United Kingdom', 21), ('Portugal', 14), ('Czech Republic', 14), ('India', 13), ('Sweden', 7), ('Spain', 7), ('Poland', 7), ('Norway', 7), ('Netherlands', 7), ('Italy', 7), ('Ireland', 7), ('Hungary', 7), ('Finland', 7), ('Denmark', 7), ('Chile', 7), ('Belgium', 7), ('Austria', 7), ('Australia', 7), ('Argentina', 7)]Relevant Documents: 1,3,4
Cited Documents: 4
Answer: The countries with the most invoices are the USA (91), Canada (56), and France (35).
Grounded answer: The countries with the most invoices are the <co: 4="">USA (91</co:>), <co: 4="">Canada (56</co:>), and <co: 4="">France (35</co:>).
> Finished chain.

1
print(output['output'])

The countries with the most invoices are the USA (91), Canada (56), and France (35).

1
output=agent_executor.invoke({
2
   "input": 'who is the best customer? The customer who has spent the most money is the best.',
3
})

> Entering new AgentExecutor chain...

I will run a SQL query to find the customer who has spent the most money.
{'tool_name': 'sql_db_query', 'parameters': {'query': 'SELECT customer_name, SUM(total_cost) AS total_spent FROM orders GROUP BY customer_name ORDER BY total_spent DESC LIMIT 1;'}}
Error: (sqlite3.OperationalError) no such table: orders
[SQL: SELECT customer_name, SUM(total_cost) AS total_spent FROM orders GROUP BY customer_name ORDER BY total_spent DESC LIMIT 1;]
(Background on this error at: https://sqlalche.me/e/20/e3q8)
I received an error message saying that there is no table called 'orders'. I will now use the sql_db_list_tables tool to find out what tables are available.
{'tool_name': 'sql_db_list_tables', 'parameters': {'tool_input': ''}}
Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track
I found that there is a table called 'Customer'. I will now use the sql_db_schema tool to find out what columns are in the 'Customer' table.
{'tool_name': 'sql_db_schema', 'parameters': {'table_names': 'Customer'}}

CREATE TABLE "Customer" (
    "CustomerId" INTEGER NOT NULL,
    "FirstName" NVARCHAR(40) NOT NULL,
    "LastName" NVARCHAR(20) NOT NULL,
    "Company" NVARCHAR(80),
    "Address" NVARCHAR(70),
    "City" NVARCHAR(40),
    "State" NVARCHAR(40),
    "Country" NVARCHAR(40),
    "PostalCode" NVARCHAR(10),
    "Phone" NVARCHAR(24),
    "Fax" NVARCHAR(24),
    "Email" NVARCHAR(60) NOT NULL,
    "SupportRepId" INTEGER,
    PRIMARY KEY ("CustomerId"),
    FOREIGN KEY("SupportRepId") REFERENCES "Employee" ("EmployeeId")
)
/*
3 rows from Customer table:
CustomerId	FirstName	LastName	Company	Address	City	State	Country	PostalCode	Phone	Fax	Email	SupportRepId
1	Luís	Gonçalves	Embraer - Empresa Brasileira de Aeronáutica S.A.	Av. Brigadeiro Faria Lima, 2170	São José dos Campos	SP	Brazil	12227-000	+55 (12) 3923-5555	+55 (12) 3923-5566	luisg@embraer.com.br	3
2	Leonie	Köhler	None	Theodor-Heuss-Straße 34	Stuttgart	None	Germany	70174	+49 0711 2842222	None	leonekohler@surfeu.de	5
3	François	Tremblay	None	1498 rue Bélanger	Montréal	QC	Canada	H2G 1A7	+1 (514) 721-4711	None	ftremblay@gmail.com	3
*/
I found that the 'Customer' table does not contain any information about how much money a customer has spent. I will now use the sql_db_schema tool to find out what columns are in the 'Invoice' table.
{'tool_name': 'sql_db_schema', 'parameters': {'table_names': 'Invoice'}}

CREATE TABLE "Invoice" (
    "InvoiceId" INTEGER NOT NULL,
    "CustomerId" INTEGER NOT NULL,
    "InvoiceDate" DATETIME NOT NULL,
    "BillingAddress" NVARCHAR(70),
    "BillingCity" NVARCHAR(40),
    "BillingState" NVARCHAR(40),
    "BillingCountry" NVARCHAR(40),
    "BillingPostalCode" NVARCHAR(10),
    "Total" NUMERIC(10, 2) NOT NULL,
    PRIMARY KEY ("InvoiceId"),
    FOREIGN KEY("CustomerId") REFERENCES "Customer" ("CustomerId")
)
/*
3 rows from Invoice table:
InvoiceId	CustomerId	InvoiceDate	BillingAddress	BillingCity	BillingState	BillingCountry	BillingPostalCode	Total
1	2	2021-01-01 00:00:00	Theodor-Heuss-Straße 34	Stuttgart	None	Germany	70174	1.98
2	4	2021-01-02 00:00:00	Ullevålsveien 14	Oslo	None	Norway	0171	3.96
3	8	2021-01-03 00:00:00	Grétrystraat 63	Brussels	None	Belgium	1000	5.94
*/
I found that the 'Invoice' table contains a 'Total' column, which is likely to be the total amount spent by the customer on that invoice. I will now use the sql_db_query tool to find the customer who has spent the most money.
{'tool_name': 'sql_db_query', 'parameters': {'query': 'SELECT c.FirstName, c.LastName, SUM(i.Total) AS total_spent FROM Customer c JOIN Invoice i ON c.CustomerId = i.CustomerId GROUP BY c.CustomerId ORDER BY total_spent DESC LIMIT 1;'}}
[('Helena', 'Holý', 49.62)]Relevant Documents: 1,2,3,4
Cited Documents: 4
Answer: The best customer is Helena Holý, who has spent a total of 49.62.
Grounded answer: The best customer is <co: 4="">Helena Holý</co:>, who has spent a total of <co: 4="">49.62</co:>.
> Finished chain.

1
print(output['output'])

The best customer is Helena Holý, who has spent a total of 49.62.

1
print('**Context to pass to LLM on tables**')
2
print('Table Names')
3
print(context['table_names'])
4
print('Table Schemas')
5
print(context['table_info'])

1
**Context to pass to LLM on tables**
2
Table Names
3
Album, Artist, Customer, Employee, Genre, Invoice, InvoiceLine, MediaType, Playlist, PlaylistTrack, Track
4
Table Schemas
5
6
CREATE TABLE "Album" (
7
    "AlbumId" INTEGER NOT NULL,
8
    "Title" NVARCHAR(160) NOT NULL,
9
    "ArtistId" INTEGER NOT NULL,
10
    PRIMARY KEY ("AlbumId"),
11
    FOREIGN KEY("ArtistId") REFERENCES "Artist" ("ArtistId")
12
)
13
14
/*
15
3 rows from Album table:
16
AlbumId	Title	ArtistId
17
1	For Those About To Rock We Salute You	1
18
2	Balls to the Wall	2
19
3	Restless and Wild	2
20
*/
21
22
23
CREATE TABLE "Artist" (
24
    "ArtistId" INTEGER NOT NULL,
25
    "Name" NVARCHAR(120),
26
    PRIMARY KEY ("ArtistId")
27
)
28
29
/*
30
3 rows from Artist table:
31
ArtistId	Name
32
1	AC/DC
33
2	Accept
34
3	Aerosmith
35
*/
36
37
38
CREATE TABLE "Customer" (
39
    "CustomerId" INTEGER NOT NULL,
40
    "FirstName" NVARCHAR(40) NOT NULL,
41
    "LastName" NVARCHAR(20) NOT NULL,
42
    "Company" NVARCHAR(80),
43
    "Address" NVARCHAR(70),
44
    "City" NVARCHAR(40),
45
    "State" NVARCHAR(40),
46
    "Country" NVARCHAR(40),
47
    "PostalCode" NVARCHAR(10),
48
    "Phone" NVARCHAR(24),
49
    "Fax" NVARCHAR(24),
50
    "Email" NVARCHAR(60) NOT NULL,
51
    "SupportRepId" INTEGER,
52
    PRIMARY KEY ("CustomerId"),
53
    FOREIGN KEY("SupportRepId") REFERENCES "Employee" ("EmployeeId")
54
)
55
56
/*
57
3 rows from Customer table:
58
CustomerId	FirstName	LastName	Company	Address	City	State	Country	PostalCode	Phone	Fax	Email	SupportRepId
59
1	Luís	Gonçalves	Embraer - Empresa Brasileira de Aeronáutica S.A.	Av. Brigadeiro Faria Lima, 2170	São José dos Campos	SP	Brazil	12227-000	+55 (12) 3923-5555	+55 (12) 3923-5566	luisg@embraer.com.br	3
60
2	Leonie	Köhler	None	Theodor-Heuss-Straße 34	Stuttgart	None	Germany	70174	+49 0711 2842222	None	leonekohler@surfeu.de	5
61
3	François	Tremblay	None	1498 rue Bélanger	Montréal	QC	Canada	H2G 1A7	+1 (514) 721-4711	None	ftremblay@gmail.com	3
62
*/
63
64
65
CREATE TABLE "Employee" (
66
    "EmployeeId" INTEGER NOT NULL,
67
    "LastName" NVARCHAR(20) NOT NULL,
68
    "FirstName" NVARCHAR(20) NOT NULL,
69
    "Title" NVARCHAR(30),
70
    "ReportsTo" INTEGER,
71
    "BirthDate" DATETIME,
72
    "HireDate" DATETIME,
73
    "Address" NVARCHAR(70),
74
    "City" NVARCHAR(40),
75
    "State" NVARCHAR(40),
76
    "Country" NVARCHAR(40),
77
    "PostalCode" NVARCHAR(10),
78
    "Phone" NVARCHAR(24),
79
    "Fax" NVARCHAR(24),
80
    "Email" NVARCHAR(60),
81
    PRIMARY KEY ("EmployeeId"),
82
    FOREIGN KEY("ReportsTo") REFERENCES "Employee" ("EmployeeId")
83
)
84
85
/*
86
3 rows from Employee table:
87
EmployeeId	LastName	FirstName	Title	ReportsTo	BirthDate	HireDate	Address	City	State	Country	PostalCode	Phone	Fax	Email
88
1	Adams	Andrew	General Manager	None	1962-02-18 00:00:00	2002-08-14 00:00:00	11120 Jasper Ave NW	Edmonton	AB	Canada	T5K 2N1	+1 (780) 428-9482	+1 (780) 428-3457	andrew@chinookcorp.com
89
2	Edwards	Nancy	Sales Manager	1	1958-12-08 00:00:00	2002-05-01 00:00:00	825 8 Ave SW	Calgary	AB	Canada	T2P 2T3	+1 (403) 262-3443	+1 (403) 262-3322	nancy@chinookcorp.com
90
3	Peacock	Jane	Sales Support Agent	2	1973-08-29 00:00:00	2002-04-01 00:00:00	1111 6 Ave SW	Calgary	AB	Canada	T2P 5M5	+1 (403) 262-3443	+1 (403) 262-6712	jane@chinookcorp.com
91
*/
92
93
94
CREATE TABLE "Genre" (
95
    "GenreId" INTEGER NOT NULL,
96
    "Name" NVARCHAR(120),
97
    PRIMARY KEY ("GenreId")
98
)
99
100
/*
101
3 rows from Genre table:
102
GenreId	Name
103
1	Rock
104
2	Jazz
105
3	Metal
106
*/
107
108
109
CREATE TABLE "Invoice" (
110
    "InvoiceId" INTEGER NOT NULL,
111
    "CustomerId" INTEGER NOT NULL,
112
    "InvoiceDate" DATETIME NOT NULL,
113
    "BillingAddress" NVARCHAR(70),
114
    "BillingCity" NVARCHAR(40),
115
    "BillingState" NVARCHAR(40),
116
    "BillingCountry" NVARCHAR(40),
117
    "BillingPostalCode" NVARCHAR(10),
118
    "Total" NUMERIC(10, 2) NOT NULL,
119
    PRIMARY KEY ("InvoiceId"),
120
    FOREIGN KEY("CustomerId") REFERENCES "Customer" ("CustomerId")
121
)
122
123
/*
124
3 rows from Invoice table:
125
InvoiceId	CustomerId	InvoiceDate	BillingAddress	BillingCity	BillingState	BillingCountry	BillingPostalCode	Total
126
1	2	2021-01-01 00:00:00	Theodor-Heuss-Straße 34	Stuttgart	None	Germany	70174	1.98
127
2	4	2021-01-02 00:00:00	Ullevålsveien 14	Oslo	None	Norway	0171	3.96
128
3	8	2021-01-03 00:00:00	Grétrystraat 63	Brussels	None	Belgium	1000	5.94
129
*/
130
131
132
CREATE TABLE "InvoiceLine" (
133
    "InvoiceLineId" INTEGER NOT NULL,
134
    "InvoiceId" INTEGER NOT NULL,
135
    "TrackId" INTEGER NOT NULL,
136
    "UnitPrice" NUMERIC(10, 2) NOT NULL,
137
    "Quantity" INTEGER NOT NULL,
138
    PRIMARY KEY ("InvoiceLineId"),
139
    FOREIGN KEY("TrackId") REFERENCES "Track" ("TrackId"),
140
    FOREIGN KEY("InvoiceId") REFERENCES "Invoice" ("InvoiceId")
141
)
142
143
/*
144
3 rows from InvoiceLine table:
145
InvoiceLineId	InvoiceId	TrackId	UnitPrice	Quantity
146
1	1	2	0.99	1
147
2	1	4	0.99	1
148
3	2	6	0.99	1
149
*/
150
151
152
CREATE TABLE "MediaType" (
153
    "MediaTypeId" INTEGER NOT NULL,
154
    "Name" NVARCHAR(120),
155
    PRIMARY KEY ("MediaTypeId")
156
)
157
158
/*
159
3 rows from MediaType table:
160
MediaTypeId	Name
161
1	MPEG audio file
162
2	Protected AAC audio file
163
3	Protected MPEG-4 video file
164
*/
165
166
167
CREATE TABLE "Playlist" (
168
    "PlaylistId" INTEGER NOT NULL,
169
    "Name" NVARCHAR(120),
170
    PRIMARY KEY ("PlaylistId")
171
)
172
173
/*
174
3 rows from Playlist table:
175
PlaylistId	Name
176
1	Music
177
2	Movies
178
3	TV Shows
179
*/
180
181
182
CREATE TABLE "PlaylistTrack" (
183
    "PlaylistId" INTEGER NOT NULL,
184
    "TrackId" INTEGER NOT NULL,
185
    PRIMARY KEY ("PlaylistId", "TrackId"),
186
    FOREIGN KEY("TrackId") REFERENCES "Track" ("TrackId"),
187
    FOREIGN KEY("PlaylistId") REFERENCES "Playlist" ("PlaylistId")
188
)
189
190
/*
191
3 rows from PlaylistTrack table:
192
PlaylistId	TrackId
193
1	3402
194
1	3389
195
1	3390
196
*/
197
198
199
CREATE TABLE "Track" (
200
    "TrackId" INTEGER NOT NULL,
201
    "Name" NVARCHAR(200) NOT NULL,
202
    "AlbumId" INTEGER,
203
    "MediaTypeId" INTEGER NOT NULL,
204
    "GenreId" INTEGER,
205
    "Composer" NVARCHAR(220),
206
    "Milliseconds" INTEGER NOT NULL,
207
    "Bytes" INTEGER,
208
    "UnitPrice" NUMERIC(10, 2) NOT NULL,
209
    PRIMARY KEY ("TrackId"),
210
    FOREIGN KEY("MediaTypeId") REFERENCES "MediaType" ("MediaTypeId"),
211
    FOREIGN KEY("GenreId") REFERENCES "Genre" ("GenreId"),
212
    FOREIGN KEY("AlbumId") REFERENCES "Album" ("AlbumId")
213
)
214
215
/*
216
3 rows from Track table:
217
TrackId	Name	AlbumId	MediaTypeId	GenreId	Composer	Milliseconds	Bytes	UnitPrice
218
1	For Those About To Rock (We Salute You)	1	1	1	Angus Young, Malcolm Young, Brian Johnson	343719	11170334	0.99
219
2	Balls to the Wall	2	2	1	U. Dirkschneider, W. Hoffmann, H. Frank, P. Baltes, S. Kaufmann, G. Hoffmann	342562	5510424	0.99
220
3	Fast As a Shark	3	2	1	F. Baltes, S. Kaufman, U. Dirkscneider & W. Hoffman	230619	3990994	0.99
221
*/

1
preamble="""## Task And Context
2
You use your advanced complex reasoning capabilities to help people by answering their questions and other requests interactively. You will be asked a very wide array of requests on all kinds of topics. You will be equipped with a wide range of search engines or similar tools to help you, which you use to research your answer. You may need to use multiple tools in parallel or sequentially to complete your task. You should focus on serving the user's needs as best you can, which will be wide-ranging.
3
4
## Style Guide
5
Unless the user asks for a different style of answer, you should answer in full sentences, using proper grammar and spelling.
6
7
## Additional Information
8
You are an expert who answers the user's question by creating SQL queries and executing them.
9
You are equipped with a number of relevant SQL tools.
10
11
Here is information about the database:
12
{schema_info}
13
""".format(schema_info=context)

1
output=agent_executor.invoke({
2
   "input": 'provide the name of the best customer? The customer who has spent the most money is the best.',
3
   "preamble": preamble
4
})

> Entering new AgentExecutor chain...

I will write a SQL query to find the customer who has spent the most money.
{'tool_name': 'sql_db_query', 'parameters': {'query': 'SELECT c.FirstName, c.LastName, SUM(i.Total) AS TotalSpent FROM Customer c JOIN Invoice i ON c.CustomerId = i.CustomerId GROUP BY c.CustomerId ORDER BY TotalSpent DESC LIMIT 1;'}}
[('Helena', 'Holý', 49.62)]Relevant Documents: 0
Cited Documents: 0
Answer: The customer who has spent the most money is Helena Holý.
Grounded answer: The customer who has spent the most money is <co: 0="">Helena Holý</co:>.
> Finished chain.

1
print(output['output'])

The customer who has spent the most money is Helena Holý.