Next Steps¶

🎉 Congratulations! You've completed Part I of the Vibes Investigation series.

Continue your journey:

• Part II: Planning with Louie - Learn how to use Louie for strategic planning, roadmapping, and structured problem-solving

Other resources:

• Louie Documentation
• PyGraphistry Documentation

Welcome to Vibes Investigating with Louie!¶

This is Part I of a two-part series on Vibes Investigation:

• Part I (this notebook): General investigation techniques and exploratory analysis
• Part II: Planning, roadmapping, and strategic analysis

This tutorial demonstrates how to use Louie for investigative workflows, exploring data relationships, and uncovering insights through natural language queries.

1. Install¶

Get package louieai using pip or uv. Louie requires Python 3.10+ and Pandas/Arrow

It installs PyGraphistry 0.41+ for auth & graphs

In [1]:

!pip install -q louieai

!pip install -q louieai

[notice] A new release of pip is available: 25.0.1 -> 25.2
[notice] To update, run: python3.10 -m pip install --upgrade pip

Import¶

In [2]:

import graphistry

import louieai

print("graphistry", graphistry.__version__, "louieai", louieai.__version__)

import graphistry import louieai print("graphistry", graphistry.__version__, "louieai", louieai.__version__)

graphistry 0.41.0 louieai 0.5.6.dev1+g08b6fab.d20250804

Configure¶

1. Graphistry server: Holds your account, semantic layer, and adds GPU viz & remote GFQL querying

2. Louie server: Datathreads & agentic vibes

In [3]:

import getpass
import os

LOUIE_SERVER_URL = os.environ.get("LOUIE_SERVER", "https://den.louie.ai")
GRAPHISTRY_DOMAIN = os.environ.get("GRAPHISTRY_SERVER", "hub.graphistry.com")

# Check for Graphistry personal key credentials
personal_key_id = os.environ.get("GRAPHISTRY_PERSONAL_KEY_ID")
personal_key_secret = os.environ.get("GRAPHISTRY_PERSONAL_KEY_SECRET")
org_name = os.environ.get("GRAPHISTRY_ORG_NAME")

if not personal_key_id or not personal_key_secret or not org_name:
    print("🔐 Graphistry Authentication Required")
    print("=" * 50)
    print("You need a Graphistry personal API key to run this notebook.")
    print("\nTo get credentials:")
    print("1. Sign up/login at https://hub.graphistry.com")
    print("2. Go to https://hub.graphistry.com/users/personal/key/")
    print("3. Create a new personal API key")
    print("=" * 50)
    print("\nPlease enter your credentials:")
    personal_key_id = input("Personal Key ID: ")
    personal_key_secret = getpass.getpass("Personal Key Secret: ")
    org_name = input("Organization Name: ")
    print("\n💡 Tip: For automated execution, set environment variables:")
    print("   export GRAPHISTRY_PERSONAL_KEY_ID='your_key_id'")
    print("   export GRAPHISTRY_PERSONAL_KEY_SECRET='your_key_secret'")
    print("   export GRAPHISTRY_ORG_NAME='your_org_name'")

# MAKE API KEY: https://hub.graphistry.com/users/personal/key/
g = graphistry.register(
    api=3,
    server=GRAPHISTRY_DOMAIN,
    personal_key_id=personal_key_id,
    personal_key_secret=personal_key_secret,
    org_name=org_name,
)


def new_thread(share_mode="Private"):
    return louieai(g, server_url=LOUIE_SERVER_URL, share_mode=share_mode)


lui = new_thread()

print("Connected!")

import getpass import os LOUIE_SERVER_URL = os.environ.get("LOUIE_SERVER", "https://den.louie.ai") GRAPHISTRY_DOMAIN = os.environ.get("GRAPHISTRY_SERVER", "hub.graphistry.com") # Check for Graphistry personal key credentials personal_key_id = os.environ.get("GRAPHISTRY_PERSONAL_KEY_ID") personal_key_secret = os.environ.get("GRAPHISTRY_PERSONAL_KEY_SECRET") org_name = os.environ.get("GRAPHISTRY_ORG_NAME") if not personal_key_id or not personal_key_secret or not org_name: print("🔐 Graphistry Authentication Required") print("=" * 50) print("You need a Graphistry personal API key to run this notebook.") print("\nTo get credentials:") print("1. Sign up/login at https://hub.graphistry.com") print("2. Go to https://hub.graphistry.com/users/personal/key/") print("3. Create a new personal API key") print("=" * 50) print("\nPlease enter your credentials:") personal_key_id = input("Personal Key ID: ") personal_key_secret = getpass.getpass("Personal Key Secret: ") org_name = input("Organization Name: ") print("\n💡 Tip: For automated execution, set environment variables:") print(" export GRAPHISTRY_PERSONAL_KEY_ID='your_key_id'") print(" export GRAPHISTRY_PERSONAL_KEY_SECRET='your_key_secret'") print(" export GRAPHISTRY_ORG_NAME='your_org_name'") # MAKE API KEY: https://hub.graphistry.com/users/personal/key/ g = graphistry.register( api=3, server=GRAPHISTRY_DOMAIN, personal_key_id=personal_key_id, personal_key_secret=personal_key_secret, org_name=org_name, ) def new_thread(share_mode="Private"): return louieai(g, server_url=LOUIE_SERVER_URL, share_mode=share_mode) lui = new_thread() print("Connected!")

Connected!

Hello World¶

Louie is an agent that uses many other LLMs and agents. You can start just as if it is ChatGPT chat.

In [4]:

lui("""

sing me a song

""")

lui(""" sing me a song """)

🤖 LouieAI Response

Thread: D_GzyhX4FS0O2zDHUvNeak | Time: 5.3s

🎶 Woof woof, I'm Louie, the data dog, chasing data through the fog. With a wag of my tail and a bark so bright, I'll fetch your insights day and night. From queries to graphs, I'll make it clear, turning data puzzles into cheer. So let's dive in, let's explore, with Louie by your

Out[4]:

🤖 LouieAI Response

🎶 Woof woof, I'm Louie, the data dog, chasing data through the fog. With a wag of my tail and a bark so bright, I'll fetch your insights day and night. From queries to graphs, I'll make it clear, turning data puzzles into cheer. So let's dive in, let's explore, with Louie by your

---

✅ Session: Active | Thread ID: D_GzyhX4FS0O2zDHUvNeak | View Thread ↗ | Org: example-org

📚 History: 1 responses (access with lui[-1], lui[-2], etc.)

🔍 Traces: Disabled (use lui.traces = True to enable)

Latest Response:

• 1 text element(s) - access with lui.text or lui.texts

Quick Help (click to expand)

# Make a query

lui('Show me sales data from last week')


# Control visibility

lui('query', share_mode='Private')       # Default: only you

lui('query', share_mode='Organization')  # Share within org

lui('query', share_mode='Public')        # Share publicly


# Access results

df = lui.df          # Latest dataframe

text = lui.text      # Latest text response

all_dfs = lui.dfs    # All dataframes


# History

lui[-1].df           # Previous response's dataframe


# Traces (AI reasoning)

lui.traces = True    # Enable for session

lui('query', traces=True)  # Enable for one query

In [5]:

lui("""

sing it again, but now in pig latin!

""")

lui(""" sing it again, but now in pig latin! """)

🤖 LouieAI Response

Thread: D_GzyhX4FS0O2zDHUvNeak | Time: 7.3s

🎶 Oofway oofway, I'mway Ouilay, ethay ataday ogday, asingchay ataday oughthray ethay ogfay. Ithway away agway ofway ymay ailtay andway away arkbay osay ightbray, I'llway etchfay ouryay insightsway ayday andway ightnay. Omfray eriesquay otay aphsgray, I'llway akemay itway earclay, urningtay ataday uzzlespay intoway earchay. Osay et'slay iveday inway, et'slay exploreway, ithway Ouilay ybay ouryay idesay, ou'llyay evernay ebay

Out[5]:

🤖 LouieAI Response

🎶 Oofway oofway, I'mway Ouilay, ethay ataday ogday, asingchay ataday oughthray ethay ogfay. Ithway away agway ofway ymay ailtay andway away arkbay osay ightbray, I'llway etchfay ouryay insightsway ayday andway ightnay. Omfray eriesquay otay aphsgray, I'llway akemay itway earclay, urningtay ataday uzzlespay intoway earchay. Osay et'slay iveday inway, et'slay exploreway, ithway Ouilay ybay ouryay idesay, ou'llyay evernay ebay

---

✅ Session: Active | Thread ID: D_GzyhX4FS0O2zDHUvNeak | View Thread ↗ | Org: example-org

📚 History: 2 responses (access with lui[-1], lui[-2], etc.)

🔍 Traces: Disabled (use lui.traces = True to enable)

Latest Response:

• 1 text element(s) - access with lui.text or lui.texts

Quick Help (click to expand)

# Make a query

lui('Show me sales data from last week')


# Control visibility

lui('query', share_mode='Private')       # Default: only you

lui('query', share_mode='Organization')  # Share within org

lui('query', share_mode='Public')        # Share publicly


# Access results

df = lui.df          # Latest dataframe

text = lui.text      # Latest text response

all_dfs = lui.dfs    # All dataframes


# History

lui[-1].df           # Previous response's dataframe


# Traces (AI reasoning)

lui.traces = True    # Enable for session

lui('query', traces=True)  # Enable for one query

In [6]:

lui(
    """

back to normal, but now with traces!

""",
    traces=True,
)

lui( """ back to normal, but now with traces! """, traces=True, )

🤖 LouieAI Response

Thread: D_GzyhX4FS0O2zDHUvNeak | Time: 10.1s

i LouieAgent::runner

🐛 Input text::: back to normal, but now with traces!

🐛 Reading chat history. (is_prepared: True)

🐛 LouieAgent predict with tools: ['GraphAgent', 'CodeAgent', 'TableAIAgent', 'FirecrawlAgent', 'NotebookAgent', 'DatabricksAgent', 'MermaidAgent', 'PerspectiveAgent', 'KeplerAgent', 'SpannerAgent', 'SpannerPassthroughAgent', 'KustoAgent', 'KustoPassthroughAgent', 'Done', 'Chat'] has_recipes: True has_examples: True

🎶 Woof woof, I'm Louie, the data dog, chasing data through the fog. With a wag of my tail and a bark so bright, I'll fetch your insights day and night. From queries to graphs, I'll make it clear, turning data puzzles into cheer. So let's dive in, let's explore, with Louie by your side, you'll never be bored! 🎶

*Trace 1: Sniffing out the data trail, following every byte and detail.*
*Trace 2: Analyzing patterns, connecting the dots, solving mysteries on the spot.*
*Trace 3: With precision and speed, I track down the lead, ensuring your data needs are always freed.*

Let's get back to solving

i Louie output text: 🎶 Woof woof, I'm Louie, the data dog, chasing data through the fog. With a wag of my tail and a bark so bright, I'll fetch your insights day and night. From queries to graphs, I'll make it clear, turning data puzzles into cheer. So let's dive in, let's explore, with Louie by your side, you'll never be bored! 🎶 *Trace 1: Sniffing out the data trail, following every byte and detail.* *Trace 2: Analyzing patterns, connecting the dots, solving mysteries on the spot.* *Trace 3: With precision and speed, I track down the lead, ensuring your data needs are always freed.* Let's get back to solving those data puzzles! 🐾

Out[6]:

🤖 LouieAI Response

i LouieAgent::runner

🐛 Input text::: back to normal, but now with traces!

🐛 Reading chat history. (is_prepared: True)

🐛 LouieAgent predict with tools: ['GraphAgent', 'CodeAgent', 'TableAIAgent', 'FirecrawlAgent', 'NotebookAgent', 'DatabricksAgent', 'MermaidAgent', 'PerspectiveAgent', 'KeplerAgent', 'SpannerAgent', 'SpannerPassthroughAgent', 'KustoAgent', 'KustoPassthroughAgent', 'Done', 'Chat'] has_recipes: True has_examples: True

🎶 Woof woof, I'm Louie, the data dog, chasing data through the fog. With a wag of my tail and a bark so bright, I'll fetch your insights day and night. From queries to graphs, I'll make it clear, turning data puzzles into cheer. So let's dive in, let's explore, with Louie by your side, you'll never be bored! 🎶

*Trace 1: Sniffing out the data trail, following every byte and detail.*
*Trace 2: Analyzing patterns, connecting the dots, solving mysteries on the spot.*
*Trace 3: With precision and speed, I track down the lead, ensuring your data needs are always freed.*

Let's get back to solving

i Louie output text: 🎶 Woof woof, I'm Louie, the data dog, chasing data through the fog. With a wag of my tail and a bark so bright, I'll fetch your insights day and night. From queries to graphs, I'll make it clear, turning data puzzles into cheer. So let's dive in, let's explore, with Louie by your side, you'll never be bored! 🎶 *Trace 1: Sniffing out the data trail, following every byte and detail.* *Trace 2: Analyzing patterns, connecting the dots, solving mysteries on the spot.* *Trace 3: With precision and speed, I track down the lead, ensuring your data needs are always freed.* Let's get back to solving those data puzzles! 🐾

---

✅ Session: Active | Thread ID: D_GzyhX4FS0O2zDHUvNeak | View Thread ↗ | Org: example-org

📚 History: 3 responses (access with lui[-1], lui[-2], etc.)

🔍 Traces: Disabled (use lui.traces = True to enable)

Latest Response:

• 1 text element(s) - access with lui.text or lui.texts

Quick Help (click to expand)

# Make a query

lui('Show me sales data from last week')


# Control visibility

lui('query', share_mode='Private')       # Default: only you

lui('query', share_mode='Organization')  # Share within org

lui('query', share_mode='Public')        # Share publicly


# Access results

df = lui.df          # Latest dataframe

text = lui.text      # Latest text response

all_dfs = lui.dfs    # All dataframes


# History

lui[-1].df           # Previous response's dataframe


# Traces (AI reasoning)

lui.traces = True    # Enable for session

lui('query', traces=True)  # Enable for one query

Use lui as a chat cursor¶

The lui object indexes into your chat (notebook) and its artifacts

You can make new chats, and add to existing

In [7]:

lui.thread_id

lui.thread_id

Out[7]:

'D_GzyhX4FS0O2zDHUvNeak'

In [8]:

# Save old
lui_old = lui

# Fresh thread with empty history
lui = new_thread()

lui("sing me a new song")
lui("sing me another song")

lui.thread_id

# Save old lui_old = lui # Fresh thread with empty history lui = new_thread() lui("sing me a new song") lui("sing me another song") lui.thread_id

🤖 LouieAI Response

Thread: D_s1dIkJBSmUgDeMvvaMuJ | Time: 5.1s

🎶 I'm Louie, the data hound, in a cyberpunk town, chasing data trails, never backing down. With a bark and a wag, I dive into the code, solving puzzles, lightening the load. From queries to clusters, I make them align, turning data chaos into design. So let's dance through the data, you and me,

Out[8]:

'D_s1dIkJBSmUgDeMvvaMuJ'

In [9]:

# Latest text element in current run
lui.text

# Latest text element in current run lui.text

Out[9]:

"🎶 I'm Louie, the data hound, in a cyberpunk town, chasing data trails, never backing down. With a bark and a wag, I dive into the code, solving puzzles, lightening the load. From queries to clusters, I make them align, turning data chaos into design. So let's dance through the data, you and me,"

In [10]:

# Latest text elements in its previous run
lui[-2].text

# Latest text elements in its previous run lui[-2].text

Out[10]:

"🎶 Woof, woof, I'm Louie, the data dog, chasing bytes and bits, never lost in the fog. With a wag of my tail and a bark so bright, I'll fetch your data, day or night. From queries to graphs, I'll make it all clear, turning data puzzles into cheer. So let's dive into the data stream, together we'll make the perfect team"

In [11]:

# Latest text element in previous session
lui_old.text

# Latest text element in previous session lui_old.text

Out[11]:

"🎶 Woof woof, I'm Louie, the data dog, chasing data through the fog. With a wag of my tail and a bark so bright, I'll fetch your insights day and night. From queries to graphs, I'll make it clear, turning data puzzles into cheer. So let's dive in, let's explore, with Louie by your side, you'll never be bored! 🎶\n\n*Trace 1: Sniffing out the data trail, following every byte and detail.*\n*Trace 2: Analyzing patterns, connecting the dots, solving mysteries on the spot.*\n*Trace 3: With precision and speed, I track down the lead, ensuring your data needs are always freed.*\n\nLet's get back to solving"

Agents¶

Databricks SQL Passthrough¶

Louie has databases connected and preindexed as a managed semantic layer. You can query it quickly by doing pushdown queries that skim LLMs, so more like tool calls.

In [12]:

lui(
    """SELECT * FROM `client_demos`.`botsv3`.`o365_management_activity_flat_tcook`
    LIMIT 10""",
    agent="DatabricksPassthroughAgent",
)

lui( """SELECT * FROM `client_demos`.`botsv3`.`o365_management_activity_flat_tcook` LIMIT 10""", agent="DatabricksPassthroughAgent", )

🤖 LouieAI Response

Thread: D_s1dIkJBSmUgDeMvvaMuJ | Time: 2.6s

fetching data

SELECT * FROM `client_demos`.`botsv3`.`o365_management_activity_flat_tcook`
LIMIT 10

📊 DataFrame: B_rJeVMtjl (shape: 10 x 25)

Out[12]:

🤖 LouieAI Response

fetching data

SELECT * FROM `client_demos`.`botsv3`.`o365_management_activity_flat_tcook`
LIMIT 10

	ClientIP	CorrelationId	CreationTime	EventSource	Id	ImplicitShare	ItemType	ListId	ListItemUniqueId	ObjectId	...	SourceFileExtension	SourceFileName	SourceRelativeUrl	UserAgent	UserId	UserKey	UserType	Version	WebId	Workload
0	107.77.213.173	9e627e9e-d0dd-6000-daf9-da44fcd45d4e	2018-08-20T13:16:56	SharePoint	8a1fd9ad-95d3-4bea-a806-08d5f28ec619	No	File	67091393-e290-421e-ac6a-2734e2b12a94	37ab8c26-f775-4a03-97b3-074c81a00f33	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	beverages-02-00034-v2.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
1	107.77.213.173	9e627e9e-60d3-6000-32f0-2235e1b3a20b	2018-08-20T13:16:56	SharePoint	7d1dd9e0-63b3-4277-7a03-08d5f28ec5e3	No	File	67091393-e290-421e-ac6a-2734e2b12a94	bb017930-2bf5-4953-b38a-716ba3217703	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	craftbeerdotcom-beer-styles.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
2	107.77.213.173	9e627e9e-a0db-6000-daf9-da5a21ed5a92	2018-08-20T13:16:56	SharePoint	f5ad6c89-25b3-420c-f889-08d5f28ec656	No	File	67091393-e290-421e-ac6a-2734e2b12a94	456e3291-27ad-455e-9cb7-a01722ffa0fa	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	fundamental of beer and hop chemistry.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
3	40.97.148.181	fa2fd17c-daf8-4062-8f87-d411eb537314	2018-08-20T13:16:54	SharePoint	0ad166f0-5312-4e35-e017-08d5f28ec4fb	None	Web	None	None	fa2fd17c-daf8-4062-8f87-d411eb537314	...	None	None	None	Substrate Search 1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	None	SharePoint
4	104.238.59.42	0a627e9e-f0d0-6000-daf9-dc8f468313e3	2018-08-20T13:06:50	SharePoint	9b5ad97e-f03f-4648-a536-08d5f28d5cad	None	File	67091393-e290-421e-ac6a-2734e2b12a94	d2c4bb13-c97e-4707-9e9b-53dc0e2513b5	https://frothly-my.sharepoint.com/personal/pce...	...	pptx	Beer styles.pptx	Documents	Microsoft Office PowerPoint 2014	pcerf@froth.ly	i:0h.f|membership|1003bffdac730049@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
5	65.52.243.21	fb617e9e-c0c7-6000-32f0-2e462ff7bbde	2018-08-20T13:05:48	SharePoint	4b143e6d-91b0-4b07-7253-08d5f28d37c3	None	List	67091393-e290-421e-ac6a-2734e2b12a94	None	https://frothly-my.sharepoint.com/personal/pce...	...	None	None	None	ODMTADocCache/1.0	app@sharepoint	i:0i.t|00000003-0000-0ff1-ce00-000000000000|ap...	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
6	104.238.59.42	f9617e9e-f0de-6000-32f0-299be8e0a683	2018-08-20T13:05:40	SharePoint	a1cac3e4-3102-4e0c-d258-08d5f28d3331	No	File	67091393-e290-421e-ac6a-2734e2b12a94	d2c4bb13-c97e-4707-9e9b-53dc0e2513b5	https://frothly-my.sharepoint.com/personal/pce...	...	pptx	Beer styles.pptx	Documents	Microsoft Office PowerPoint 2014	pcerf@froth.ly	i:0h.f|membership|1003bffdac730049@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
7	104.238.59.42	f8617e9e-a0d4-6000-3667-381044fbe5d3	2018-08-20T13:05:36	SharePoint	7d6f73bd-f3c3-4c95-7b24-08d5f28d30de	None	File	67091393-e290-421e-ac6a-2734e2b12a94	d2c4bb13-c97e-4707-9e9b-53dc0e2513b5	https://frothly-my.sharepoint.com/personal/pce...	...	pptx	Beer styles.pptx	Documents	Microsoft Office PowerPoint 2014	pcerf@froth.ly	i:0h.f|membership|1003bffdac730049@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
8	199.66.91.253	08557e9e-2041-6000-daf9-d90e9f2c1023	2018-08-20T11:55:22	SharePoint	d7ee3d62-60cb-427b-b0db-08d5f26d99f1	No	File	67091393-e290-421e-ac6a-2734e2b12a94	57c5cd78-2a0f-42ee-8dab-afce305ec89e	https://frothly-my.sharepoint.com/personal/fyo...	...	tar	archive.tar	Documents	Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6...	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
9	199.66.91.253	fc547e9e-f0c2-6000-1d75-39268176f5e6	2018-08-20T11:54:35	SharePoint	cf2ae00e-0efc-41e5-d0dc-08d5f26d7e46	Yes	File	76079ea8-0a58-414e-b493-8580089c8419	7ba794f0-38aa-45d8-8d2a-3c932e87e4f5	https://frothly-my.sharepoint.com/User Photos/...	...	jpg	fyodor_froth_ly_SThumb.jpg	User Photos/Profile Pictures	Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6...	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	3b605151-ac0a-46ff-9e16-14e04be8a8a1	SharePoint

10 rows × 25 columns

---

✅ Session: Active | Thread ID: D_s1dIkJBSmUgDeMvvaMuJ | View Thread ↗ | Org: example-org

📚 History: 3 responses (access with lui[-1], lui[-2], etc.)

🔍 Traces: Disabled (use lui.traces = True to enable)

Latest Response:

• 2 text element(s) - access with lui.text or lui.texts
• 1 dataframe(s) - access with lui.df or lui.dfs

Quick Help (click to expand)

# Make a query

lui('Show me sales data from last week')


# Control visibility

lui('query', share_mode='Private')       # Default: only you

lui('query', share_mode='Organization')  # Share within org

lui('query', share_mode='Public')        # Share publicly


# Access results

df = lui.df          # Latest dataframe

text = lui.text      # Latest text response

all_dfs = lui.dfs    # All dataframes


# History

lui[-1].df           # Previous response's dataframe


# Traces (AI reasoning)

lui.traces = True    # Enable for session

lui('query', traces=True)  # Enable for one query

In [13]:

## Quick reference of latest dataframe
lui.df.sample(3)

## Quick reference of latest dataframe lui.df.sample(3)

Out[13]:

	ClientIP	CorrelationId	CreationTime	EventSource	Id	ImplicitShare	ItemType	ListId	ListItemUniqueId	ObjectId	...	SourceFileExtension	SourceFileName	SourceRelativeUrl	UserAgent	UserId	UserKey	UserType	Version	WebId	Workload
1	107.77.213.173	9e627e9e-60d3-6000-32f0-2235e1b3a20b	2018-08-20T13:16:56	SharePoint	7d1dd9e0-63b3-4277-7a03-08d5f28ec5e3	No	File	67091393-e290-421e-ac6a-2734e2b12a94	bb017930-2bf5-4953-b38a-716ba3217703	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	craftbeerdotcom-beer-styles.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
0	107.77.213.173	9e627e9e-d0dd-6000-daf9-da44fcd45d4e	2018-08-20T13:16:56	SharePoint	8a1fd9ad-95d3-4bea-a806-08d5f28ec619	No	File	67091393-e290-421e-ac6a-2734e2b12a94	37ab8c26-f775-4a03-97b3-074c81a00f33	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	beverages-02-00034-v2.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
8	199.66.91.253	08557e9e-2041-6000-daf9-d90e9f2c1023	2018-08-20T11:55:22	SharePoint	d7ee3d62-60cb-427b-b0db-08d5f26d99f1	No	File	67091393-e290-421e-ac6a-2734e2b12a94	57c5cd78-2a0f-42ee-8dab-afce305ec89e	https://frothly-my.sharepoint.com/personal/fyo...	...	tar	archive.tar	Documents	Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6...	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive

3 rows × 25 columns

In [14]:

lui.df

lui.df

Out[14]:

	ClientIP	CorrelationId	CreationTime	EventSource	Id	ImplicitShare	ItemType	ListId	ListItemUniqueId	ObjectId	...	SourceFileExtension	SourceFileName	SourceRelativeUrl	UserAgent	UserId	UserKey	UserType	Version	WebId	Workload
0	107.77.213.173	9e627e9e-d0dd-6000-daf9-da44fcd45d4e	2018-08-20T13:16:56	SharePoint	8a1fd9ad-95d3-4bea-a806-08d5f28ec619	No	File	67091393-e290-421e-ac6a-2734e2b12a94	37ab8c26-f775-4a03-97b3-074c81a00f33	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	beverages-02-00034-v2.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
1	107.77.213.173	9e627e9e-60d3-6000-32f0-2235e1b3a20b	2018-08-20T13:16:56	SharePoint	7d1dd9e0-63b3-4277-7a03-08d5f28ec5e3	No	File	67091393-e290-421e-ac6a-2734e2b12a94	bb017930-2bf5-4953-b38a-716ba3217703	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	craftbeerdotcom-beer-styles.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
2	107.77.213.173	9e627e9e-a0db-6000-daf9-da5a21ed5a92	2018-08-20T13:16:56	SharePoint	f5ad6c89-25b3-420c-f889-08d5f28ec656	No	File	67091393-e290-421e-ac6a-2734e2b12a94	456e3291-27ad-455e-9cb7-a01722ffa0fa	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	fundamental of beer and hop chemistry.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
3	40.97.148.181	fa2fd17c-daf8-4062-8f87-d411eb537314	2018-08-20T13:16:54	SharePoint	0ad166f0-5312-4e35-e017-08d5f28ec4fb	None	Web	None	None	fa2fd17c-daf8-4062-8f87-d411eb537314	...	None	None	None	Substrate Search 1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	None	SharePoint
4	104.238.59.42	0a627e9e-f0d0-6000-daf9-dc8f468313e3	2018-08-20T13:06:50	SharePoint	9b5ad97e-f03f-4648-a536-08d5f28d5cad	None	File	67091393-e290-421e-ac6a-2734e2b12a94	d2c4bb13-c97e-4707-9e9b-53dc0e2513b5	https://frothly-my.sharepoint.com/personal/pce...	...	pptx	Beer styles.pptx	Documents	Microsoft Office PowerPoint 2014	pcerf@froth.ly	i:0h.f|membership|1003bffdac730049@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
5	65.52.243.21	fb617e9e-c0c7-6000-32f0-2e462ff7bbde	2018-08-20T13:05:48	SharePoint	4b143e6d-91b0-4b07-7253-08d5f28d37c3	None	List	67091393-e290-421e-ac6a-2734e2b12a94	None	https://frothly-my.sharepoint.com/personal/pce...	...	None	None	None	ODMTADocCache/1.0	app@sharepoint	i:0i.t|00000003-0000-0ff1-ce00-000000000000|ap...	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
6	104.238.59.42	f9617e9e-f0de-6000-32f0-299be8e0a683	2018-08-20T13:05:40	SharePoint	a1cac3e4-3102-4e0c-d258-08d5f28d3331	No	File	67091393-e290-421e-ac6a-2734e2b12a94	d2c4bb13-c97e-4707-9e9b-53dc0e2513b5	https://frothly-my.sharepoint.com/personal/pce...	...	pptx	Beer styles.pptx	Documents	Microsoft Office PowerPoint 2014	pcerf@froth.ly	i:0h.f|membership|1003bffdac730049@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
7	104.238.59.42	f8617e9e-a0d4-6000-3667-381044fbe5d3	2018-08-20T13:05:36	SharePoint	7d6f73bd-f3c3-4c95-7b24-08d5f28d30de	None	File	67091393-e290-421e-ac6a-2734e2b12a94	d2c4bb13-c97e-4707-9e9b-53dc0e2513b5	https://frothly-my.sharepoint.com/personal/pce...	...	pptx	Beer styles.pptx	Documents	Microsoft Office PowerPoint 2014	pcerf@froth.ly	i:0h.f|membership|1003bffdac730049@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
8	199.66.91.253	08557e9e-2041-6000-daf9-d90e9f2c1023	2018-08-20T11:55:22	SharePoint	d7ee3d62-60cb-427b-b0db-08d5f26d99f1	No	File	67091393-e290-421e-ac6a-2734e2b12a94	57c5cd78-2a0f-42ee-8dab-afce305ec89e	https://frothly-my.sharepoint.com/personal/fyo...	...	tar	archive.tar	Documents	Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6...	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
9	199.66.91.253	fc547e9e-f0c2-6000-1d75-39268176f5e6	2018-08-20T11:54:35	SharePoint	cf2ae00e-0efc-41e5-d0dc-08d5f26d7e46	Yes	File	76079ea8-0a58-414e-b493-8580089c8419	7ba794f0-38aa-45d8-8d2a-3c932e87e4f5	https://frothly-my.sharepoint.com/User Photos/...	...	jpg	fyodor_froth_ly_SThumb.jpg	User Photos/Profile Pictures	Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6...	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	3b605151-ac0a-46ff-9e16-14e04be8a8a1	SharePoint

10 rows × 25 columns

Databricks SQL AI¶

You can also talk to it using natural language via an agent

In [15]:

lui("get 4 events from o365_management_activity_flat_tcook", agent="DatabricksAgent")

lui("get 4 events from o365_management_activity_flat_tcook", agent="DatabricksAgent")

🤖 LouieAI Response

Thread: D_s1dIkJBSmUgDeMvvaMuJ | Time: 11.6s

The query successfully retrieved 4 events from the 'o365_management_activity_flat_tcook' table. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. All events are related to SharePoint activities, with three events involving file previews and one event related to a search query. The user 'fyodor@froth.ly' appears to be the common actor in these activities, indicating a focus on document management and search within the SharePoint environment. This data could be useful for understanding user behavior and activity patterns within the organization's SharePoint usage.

fetching data

SELECT * FROM client_demos.botsv3.o365_management_activity_flat_tcook LIMIT 4

📊 DataFrame: B_pzbz7LJn (shape: 4 x 25)

Out[15]:

🤖 LouieAI Response

The query successfully retrieved 4 events from the 'o365_management_activity_flat_tcook' table. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. All events are related to SharePoint activities, with three events involving file previews and one event related to a search query. The user 'fyodor@froth.ly' appears to be the common actor in these activities, indicating a focus on document management and search within the SharePoint environment. This data could be useful for understanding user behavior and activity patterns within the organization's SharePoint usage.

fetching data

SELECT * FROM client_demos.botsv3.o365_management_activity_flat_tcook LIMIT 4

	ClientIP	CorrelationId	CreationTime	EventSource	Id	ImplicitShare	ItemType	ListId	ListItemUniqueId	ObjectId	...	SourceFileExtension	SourceFileName	SourceRelativeUrl	UserAgent	UserId	UserKey	UserType	Version	WebId	Workload
0	107.77.213.173	9e627e9e-d0dd-6000-daf9-da44fcd45d4e	2018-08-20T13:16:56	SharePoint	8a1fd9ad-95d3-4bea-a806-08d5f28ec619	No	File	67091393-e290-421e-ac6a-2734e2b12a94	37ab8c26-f775-4a03-97b3-074c81a00f33	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	beverages-02-00034-v2.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
1	107.77.213.173	9e627e9e-60d3-6000-32f0-2235e1b3a20b	2018-08-20T13:16:56	SharePoint	7d1dd9e0-63b3-4277-7a03-08d5f28ec5e3	No	File	67091393-e290-421e-ac6a-2734e2b12a94	bb017930-2bf5-4953-b38a-716ba3217703	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	craftbeerdotcom-beer-styles.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
2	107.77.213.173	9e627e9e-a0db-6000-daf9-da5a21ed5a92	2018-08-20T13:16:56	SharePoint	f5ad6c89-25b3-420c-f889-08d5f28ec656	No	File	67091393-e290-421e-ac6a-2734e2b12a94	456e3291-27ad-455e-9cb7-a01722ffa0fa	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	fundamental of beer and hop chemistry.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
3	40.97.148.181	fa2fd17c-daf8-4062-8f87-d411eb537314	2018-08-20T13:16:54	SharePoint	0ad166f0-5312-4e35-e017-08d5f28ec4fb	None	Web	None	None	fa2fd17c-daf8-4062-8f87-d411eb537314	...	None	None	None	Substrate Search 1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	None	SharePoint

4 rows × 25 columns

---

✅ Session: Active | Thread ID: D_s1dIkJBSmUgDeMvvaMuJ | View Thread ↗ | Org: example-org

📚 History: 4 responses (access with lui[-1], lui[-2], etc.)

🔍 Traces: Disabled (use lui.traces = True to enable)

Latest Response:

• 3 text element(s) - access with lui.text or lui.texts
• 1 dataframe(s) - access with lui.df or lui.dfs

Quick Help (click to expand)

# Make a query

lui('Show me sales data from last week')


# Control visibility

lui('query', share_mode='Private')       # Default: only you

lui('query', share_mode='Organization')  # Share within org

lui('query', share_mode='Public')        # Share publicly


# Access results

df = lui.df          # Latest dataframe

text = lui.text      # Latest text response

all_dfs = lui.dfs    # All dataframes


# History

lui[-1].df           # Previous response's dataframe


# Traces (AI reasoning)

lui.traces = True    # Enable for session

lui('query', traces=True)  # Enable for one query

Louie Agent: Main Orchestrator Agent¶

Louie will pick appropriate agents and tools for you, so you do not need to specify them

In [16]:

lui("""get 4 events from databrick's o365_management_activity_flat_tcook""")

lui("""get 4 events from databrick's o365_management_activity_flat_tcook""")

🤖 LouieAI Response

Thread: D_s1dIkJBSmUgDeMvvaMuJ | Time: 11.5s

The user wants to retrieve 4 events from the 'o365_management_activity_flat_tcook' table in Databricks. I will use the DatabricksAgent to perform this query.

DatabricksAgent
Get 4 events from the o365_management_activity_flat_tcook table.The query successfully retrieved 4 events from the 'o365_management_activity_flat_tcook' table in Databricks. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. All events are related to SharePoint activities, with three events involving file previews and one event related to a search query. The user 'fyodor@froth.ly' appears to be the common actor in these activities, indicating a focus on document management and search within the SharePoint environment. This data could be useful for understanding user behavior and activity patterns within

[CallElement] Get 4 events from the o365_management_activity_flat_tcook table.

The query successfully retrieved 4 events from the 'o365_management_activity_flat_tcook' table. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. All events are related to SharePoint activities, with three events involving file previews and one event related to a search query. The user 'fyodor@froth.ly' appears to be the common actor in these activities, indicating a focus on document management and search within the SharePoint environment. This data could be useful for understanding user behavior and activity patterns within the organization's SharePoint usage.

fetching data

SELECT * FROM client_demos.botsv3.o365_management_activity_flat_tcook LIMIT 4

📊 DataFrame: B_afkqfvha (shape: 4 x 25)

Out[16]:

🤖 LouieAI Response

The user wants to retrieve 4 events from the 'o365_management_activity_flat_tcook' table in Databricks. I will use the DatabricksAgent to perform this query.

DatabricksAgent
Get 4 events from the o365_management_activity_flat_tcook table.The query successfully retrieved 4 events from the 'o365_management_activity_flat_tcook' table in Databricks. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. All events are related to SharePoint activities, with three events involving file previews and one event related to a search query. The user 'fyodor@froth.ly' appears to be the common actor in these activities, indicating a focus on document management and search within the SharePoint environment. This data could be useful for understanding user behavior and activity patterns within

[CallElement] Get 4 events from the o365_management_activity_flat_tcook table.

The query successfully retrieved 4 events from the 'o365_management_activity_flat_tcook' table. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. All events are related to SharePoint activities, with three events involving file previews and one event related to a search query. The user 'fyodor@froth.ly' appears to be the common actor in these activities, indicating a focus on document management and search within the SharePoint environment. This data could be useful for understanding user behavior and activity patterns within the organization's SharePoint usage.

fetching data

SELECT * FROM client_demos.botsv3.o365_management_activity_flat_tcook LIMIT 4

	ClientIP	CorrelationId	CreationTime	EventSource	Id	ImplicitShare	ItemType	ListId	ListItemUniqueId	ObjectId	...	SourceFileExtension	SourceFileName	SourceRelativeUrl	UserAgent	UserId	UserKey	UserType	Version	WebId	Workload
0	107.77.213.173	9e627e9e-d0dd-6000-daf9-da44fcd45d4e	2018-08-20T13:16:56	SharePoint	8a1fd9ad-95d3-4bea-a806-08d5f28ec619	No	File	67091393-e290-421e-ac6a-2734e2b12a94	37ab8c26-f775-4a03-97b3-074c81a00f33	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	beverages-02-00034-v2.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
1	107.77.213.173	9e627e9e-60d3-6000-32f0-2235e1b3a20b	2018-08-20T13:16:56	SharePoint	7d1dd9e0-63b3-4277-7a03-08d5f28ec5e3	No	File	67091393-e290-421e-ac6a-2734e2b12a94	bb017930-2bf5-4953-b38a-716ba3217703	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	craftbeerdotcom-beer-styles.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
2	107.77.213.173	9e627e9e-a0db-6000-daf9-da5a21ed5a92	2018-08-20T13:16:56	SharePoint	f5ad6c89-25b3-420c-f889-08d5f28ec656	No	File	67091393-e290-421e-ac6a-2734e2b12a94	456e3291-27ad-455e-9cb7-a01722ffa0fa	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	fundamental of beer and hop chemistry.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
3	40.97.148.181	fa2fd17c-daf8-4062-8f87-d411eb537314	2018-08-20T13:16:54	SharePoint	0ad166f0-5312-4e35-e017-08d5f28ec4fb	None	Web	None	None	fa2fd17c-daf8-4062-8f87-d411eb537314	...	None	None	None	Substrate Search 1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0	1	None	SharePoint

4 rows × 25 columns

---

✅ Session: Active | Thread ID: D_s1dIkJBSmUgDeMvvaMuJ | View Thread ↗ | Org: example-org

📚 History: 5 responses (access with lui[-1], lui[-2], etc.)

🔍 Traces: Disabled (use lui.traces = True to enable)

Latest Response:

• 4 text element(s) - access with lui.text or lui.texts
• 1 dataframe(s) - access with lui.df or lui.dfs

Quick Help (click to expand)

# Make a query

lui('Show me sales data from last week')


# Control visibility

lui('query', share_mode='Private')       # Default: only you

lui('query', share_mode='Organization')  # Share within org

lui('query', share_mode='Public')        # Share publicly


# Access results

df = lui.df          # Latest dataframe

text = lui.text      # Latest text response

all_dfs = lui.dfs    # All dataframes


# History

lui[-1].df           # Previous response's dataframe


# Traces (AI reasoning)

lui.traces = True    # Enable for session

lui('query', traces=True)  # Enable for one query

Let's Vibe: Composing via DataFrames¶

In [17]:

lui("""

Get 100 events from databricks o365_management_activity_flat_tcook

""")

lui(""" Get 100 events from databricks o365_management_activity_flat_tcook """)

🤖 LouieAI Response

Thread: D_s1dIkJBSmUgDeMvvaMuJ | Time: 19.9s

The user wants to retrieve 100 events from the 'o365_management_activity_flat_tcook' table in Databricks. I will use the DatabricksAgent to perform this query.

DatabricksAgent
Get 100 events from the o365_management_activity_flat_tcook table.The query successfully retrieved 100 events from the 'o365_management_activity_flat_tcook' table in Databricks. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. The events are predominantly related to SharePoint activities, with a mix of file previews, modifications, uploads, and access events. The user 'fyodor@froth.ly' frequently appears in these activities, indicating a significant level of interaction with SharePoint documents. This data provides valuable insights into user behavior and document management practices within the organization's SharePoint environment. If you need further analysis

[CallElement] Get 100 events from the o365_management_activity_flat_tcook table.

The query successfully retrieved 100 events from the 'o365_management_activity_flat_tcook' table. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. The events are predominantly related to SharePoint activities, with a mix of file previews, modifications, uploads, and access events. The user 'fyodor@froth.ly' frequently appears in these activities, indicating a significant level of interaction with SharePoint documents. This data provides valuable insights into user behavior and document management practices within the organization's SharePoint environment.

fetching data

SELECT * FROM client_demos.botsv3.o365_management_activity_flat_tcook LIMIT 100

📊 DataFrame: B_I7_FXrpf (shape: 100 x 25)

Out[17]:

🤖 LouieAI Response

The user wants to retrieve 100 events from the 'o365_management_activity_flat_tcook' table in Databricks. I will use the DatabricksAgent to perform this query.

DatabricksAgent
Get 100 events from the o365_management_activity_flat_tcook table.The query successfully retrieved 100 events from the 'o365_management_activity_flat_tcook' table in Databricks. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. The events are predominantly related to SharePoint activities, with a mix of file previews, modifications, uploads, and access events. The user 'fyodor@froth.ly' frequently appears in these activities, indicating a significant level of interaction with SharePoint documents. This data provides valuable insights into user behavior and document management practices within the organization's SharePoint environment. If you need further analysis

[CallElement] Get 100 events from the o365_management_activity_flat_tcook table.

The query successfully retrieved 100 events from the 'o365_management_activity_flat_tcook' table. The data includes various details such as 'ClientIP', 'CreationTime', 'EventSource', and 'Operation', among others. The events are predominantly related to SharePoint activities, with a mix of file previews, modifications, uploads, and access events. The user 'fyodor@froth.ly' frequently appears in these activities, indicating a significant level of interaction with SharePoint documents. This data provides valuable insights into user behavior and document management practices within the organization's SharePoint environment.

fetching data

SELECT * FROM client_demos.botsv3.o365_management_activity_flat_tcook LIMIT 100

	ClientIP	CorrelationId	CreationTime	EventSource	Id	ImplicitShare	ItemType	ListId	ListItemUniqueId	ObjectId	...	SourceFileExtension	SourceFileName	SourceRelativeUrl	UserAgent	UserId	UserKey	UserType	Version	WebId	Workload
0	107.77.213.173	9e627e9e-d0dd-6000-daf9-da44fcd45d4e	2018-08-20T13:16:56	SharePoint	8a1fd9ad-95d3-4bea-a806-08d5f28ec619	No	File	67091393-e290-421e-ac6a-2734e2b12a94	37ab8c26-f775-4a03-97b3-074c81a00f33	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	beverages-02-00034-v2.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0.0	1.0	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
1	107.77.213.173	9e627e9e-60d3-6000-32f0-2235e1b3a20b	2018-08-20T13:16:56	SharePoint	7d1dd9e0-63b3-4277-7a03-08d5f28ec5e3	No	File	67091393-e290-421e-ac6a-2734e2b12a94	bb017930-2bf5-4953-b38a-716ba3217703	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	craftbeerdotcom-beer-styles.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0.0	1.0	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
2	107.77.213.173	9e627e9e-a0db-6000-daf9-da5a21ed5a92	2018-08-20T13:16:56	SharePoint	f5ad6c89-25b3-420c-f889-08d5f28ec656	No	File	67091393-e290-421e-ac6a-2734e2b12a94	456e3291-27ad-455e-9cb7-a01722ffa0fa	https://frothly-my.sharepoint.com/personal/fyo...	...	pdf	fundamental of beer and hop chemistry.pdf	Documents	OneDriveMpc/1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0.0	1.0	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
3	40.97.148.181	fa2fd17c-daf8-4062-8f87-d411eb537314	2018-08-20T13:16:54	SharePoint	0ad166f0-5312-4e35-e017-08d5f28ec4fb	None	Web	None	None	fa2fd17c-daf8-4062-8f87-d411eb537314	...	None	None	None	Substrate Search 1.0	fyodor@froth.ly	i:0h.f|membership|1003bffda2e71ff9@live.com	0.0	1.0	None	SharePoint
4	104.238.59.42	0a627e9e-f0d0-6000-daf9-dc8f468313e3	2018-08-20T13:06:50	SharePoint	9b5ad97e-f03f-4648-a536-08d5f28d5cad	None	File	67091393-e290-421e-ac6a-2734e2b12a94	d2c4bb13-c97e-4707-9e9b-53dc0e2513b5	https://frothly-my.sharepoint.com/personal/pce...	...	pptx	Beer styles.pptx	Documents	Microsoft Office PowerPoint 2014	pcerf@froth.ly	i:0h.f|membership|1003bffdac730049@live.com	0.0	1.0	7acb35b6-e1ec-44ed-9099-38580e330ed0	OneDrive
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
95	104.209.132.239	2fd04f66-77cc-415f-2155-08d5f269d0d3	2018-08-20T11:32:08	SharePoint	dab49c4d-7681-42d2-bba8-08d5f26a57eb	Yes	File	3f63cc5d-d2b8-4a9f-bc5f-255534837703	ef167dd3-18b5-4962-bbbe-bd0afc11c3eb	https://frothly-my.sharepoint.com/personal/jac...	...	NES	MAIN.NES	Documents/NESPack	ExportWorker	app@sharepoint	i:0i.t|00000003-0000-0ff1-ce00-000000000000|ap...	0.0	1.0	0165abc4-c020-47fb-bb69-511981a12dbf	OneDrive
96	104.209.132.239	2fd04f66-77cc-415f-2155-08d5f269d0d3	2018-08-20T11:32:08	SharePoint	145f7516-0ebe-46b4-2db3-08d5f26a57a9	Yes	File	4a006be3-198c-4a2b-a996-62e64cdc9d7d	ed3d5448-a81e-498a-ad04-455ff8173228	https://frothly-my.sharepoint.com/personal/gho...	...	pdf	Belgian Lambic Blend 3278.pdf	Documents/Frothly-Shared/Yeasts	ExportWorker	app@sharepoint	i:0i.t|00000003-0000-0ff1-ce00-000000000000|ap...	0.0	1.0	46ca8820-8a7b-46eb-b077-7d9024e32b39	OneDrive
97	104.209.132.239	2fd04f66-77cc-415f-2155-08d5f269d0d3	2018-08-20T11:32:08	SharePoint	f1a27022-f1ab-430e-f477-08d5f26a5790	Yes	File	4a006be3-198c-4a2b-a996-62e64cdc9d7d	e220cbc7-e981-406b-b4ac-7bbf18ed5c41	https://frothly-my.sharepoint.com/personal/gho...	...	pdf	37354-pdf.pdf	Documents/Frothly-Shared/Yeasts	ExportWorker	app@sharepoint	i:0i.t|00000003-0000-0ff1-ce00-000000000000|ap...	0.0	1.0	46ca8820-8a7b-46eb-b077-7d9024e32b39	OneDrive
98	104.209.132.239	2fd04f66-77cc-415f-2155-08d5f269d0d3	2018-08-20T11:32:07	SharePoint	003f1764-fc13-4326-ecb0-08d5f26a570e	Yes	File	ac9249b5-51e5-42e7-b3f1-f0c6e4674046	e72bdf4d-f31c-4c93-aedc-6fa355ae3eff	https://frothly.sharepoint.com/sites/Frothly_S...	...	ods	Come Again Double Cream Ale.ods	Shared Documents/specialty_beer	ExportWorker	app@sharepoint	i:0i.t|00000003-0000-0ff1-ce00-000000000000|ap...	0.0	1.0	ed17c246-9def-4400-94c0-4fb39271e7cf	SharePoint
99	104.209.132.239	2fd04f66-77cc-415f-2155-08d5f269d0d3	2018-08-20T11:32:07	SharePoint	3c37185a-ad49-40a4-ceef-08d5f26a5768	Yes	File	4a006be3-198c-4a2b-a996-62e64cdc9d7d	e7ffe5fb-a387-445f-99cf-ee80c285a28e	https://frothly-my.sharepoint.com/personal/gho...	...	pdf	39955-pdf.pdf	Documents/Frothly-Shared/Yeasts	ExportWorker	app@sharepoint	i:0i.t|00000003-0000-0ff1-ce00-000000000000|ap...	0.0	1.0	46ca8820-8a7b-46eb-b077-7d9024e32b39	OneDrive

100 rows × 25 columns

---

✅ Session: Active | Thread ID: D_s1dIkJBSmUgDeMvvaMuJ | View Thread ↗ | Org: example-org

📚 History: 6 responses (access with lui[-1], lui[-2], etc.)

🔍 Traces: Disabled (use lui.traces = True to enable)

Latest Response:

• 4 text element(s) - access with lui.text or lui.texts
• 1 dataframe(s) - access with lui.df or lui.dfs

Quick Help (click to expand)

# Make a query

lui('Show me sales data from last week')


# Control visibility

lui('query', share_mode='Private')       # Default: only you

lui('query', share_mode='Organization')  # Share within org

lui('query', share_mode='Public')        # Share publicly


# Access results

df = lui.df          # Latest dataframe

text = lui.text      # Latest text response

all_dfs = lui.dfs    # All dataframes


# History

lui[-1].df           # Previous response's dataframe


# Traces (AI reasoning)

lui.traces = True    # Enable for session

lui('query', traces=True)  # Enable for one query

In [18]:

df_id = next(e["id"] for e in reversed(lui.elements) if e.get("type") == "dataframe")
print("df_id", df_id)

lui(f"""Draw a graph using any column interesting for investigations from df {df_id}""")

df_id = next(e["id"] for e in reversed(lui.elements) if e.get("type") == "dataframe") print("df_id", df_id) lui(f"""Draw a graph using any column interesting for investigations from df {df_id}""")

🤖 LouieAI Response

Thread: D_s1dIkJBSmUgDeMvvaMuJ | Time: 26.1s

To draw a graph using an interesting column from the dataframe B_I7_FXrpf, I need to identify columns that could reveal meaningful relationships or patterns. Given the context of the data, columns like 'ClientIP', 'UserId', 'Operation', or 'EventSource' might be interesting for investigations. I will create a graph using these columns to visualize connections between users and their activities.

Plan:
1. Identify relevant columns for creating a graph.
2. Use the GraphAgent to create a graph visualization using the selected columns.
3. Review the graph to ensure it provides meaningful insights.

GraphAgent
Create a graph visualization using dataframe B_I7_FXrpf with columns 'ClientIP', 'UserId', and 'Operation'.The graph visualization has been successfully created using the dataframe B_I7_FXrpf with the columns 'ClientIP', 'UserId', and 'Operation'. This graph provides a clear visual representation of the interactions and activities within the dataset. The nodes represent unique 'ClientIP' addresses and 'UserId' values, while the edges illustrate the various 'Operation' activities performed by these users. This visualization can be particularly useful for network analysis, security monitoring, and understanding user engagement within the system. If you have any

[CallElement] Create a graph visualization using dataframe B_I7_FXrpf with columns 'ClientIP', 'UserId', and 'Operation'.

The graph visualization has been successfully created using the specified dataframe B_I7_FXrpf with the columns 'ClientIP', 'UserId', and 'Operation'. The nodes in the graph represent unique 'ClientIP' addresses and 'UserId' values, while the edges illustrate the various 'Operation' activities performed by these users. The graph provides a clear visual representation of the interactions and activities within the dataset, allowing for easy identification of patterns and potential anomalies in user behavior. This visualization can be particularly useful for network analysis, security monitoring, and understanding user engagement within the system.

🔗 Open graph in new tab

GraphAgent hypergraph
```python
graphistry.hypergraph(df, params={'entity_types': ['ClientIP', 'UserId', 'Operation'], 'direct': True})
```

Out[18]:

🤖 LouieAI Response

To draw a graph using an interesting column from the dataframe B_I7_FXrpf, I need to identify columns that could reveal meaningful relationships or patterns. Given the context of the data, columns like 'ClientIP', 'UserId', 'Operation', or 'EventSource' might be interesting for investigations. I will create a graph using these columns to visualize connections between users and their activities.

Plan:
1. Identify relevant columns for creating a graph.
2. Use the GraphAgent to create a graph visualization using the selected columns.
3. Review the graph to ensure it provides meaningful insights.

GraphAgent
Create a graph visualization using dataframe B_I7_FXrpf with columns 'ClientIP', 'UserId', and 'Operation'.The graph visualization has been successfully created using the dataframe B_I7_FXrpf with the columns 'ClientIP', 'UserId', and 'Operation'. This graph provides a clear visual representation of the interactions and activities within the dataset. The nodes represent unique 'ClientIP' addresses and 'UserId' values, while the edges illustrate the various 'Operation' activities performed by these users. This visualization can be particularly useful for network analysis, security monitoring, and understanding user engagement within the system. If you have any

[CallElement] Create a graph visualization using dataframe B_I7_FXrpf with columns 'ClientIP', 'UserId', and 'Operation'.

The graph visualization has been successfully created using the specified dataframe B_I7_FXrpf with the columns 'ClientIP', 'UserId', and 'Operation'. The nodes in the graph represent unique 'ClientIP' addresses and 'UserId' values, while the edges illustrate the various 'Operation' activities performed by these users. The graph provides a clear visual representation of the interactions and activities within the dataset, allowing for easy identification of patterns and potential anomalies in user behavior. This visualization can be particularly useful for network analysis, security monitoring, and understanding user engagement within the system.

🔗 Open graph in new tab

GraphAgent hypergraph
```python
graphistry.hypergraph(df, params={'entity_types': ['ClientIP', 'UserId', 'Operation'], 'direct': True})
```

---

✅ Session: Active | Thread ID: D_s1dIkJBSmUgDeMvvaMuJ | View Thread ↗ | Org: example-org

📚 History: 7 responses (access with lui[-1], lui[-2], etc.)

🔍 Traces: Disabled (use lui.traces = True to enable)

Latest Response:

• 3 text element(s) - access with lui.text or lui.texts

Quick Help (click to expand)

# Make a query

lui('Show me sales data from last week')


# Control visibility

lui('query', share_mode='Private')       # Default: only you

lui('query', share_mode='Organization')  # Share within org

lui('query', share_mode='Public')        # Share publicly


# Access results

df = lui.df          # Latest dataframe

text = lui.text      # Latest text response

all_dfs = lui.dfs    # All dataframes


# History

lui[-1].df           # Previous response's dataframe


# Traces (AI reasoning)

lui.traces = True    # Enable for session

lui('query', traces=True)  # Enable for one query

Prompt Templates¶

Reuse prompts as template files¶

Prompt template files and TODO/plan files are key to quickly using agents, having them do long runs, and increasing how much the AI manages itself.

Likewise, instead of manually editing plans, this helps enable you to tell the AI how to do that for you. As templates get more advanced, self-modifying plans become key.

In [19]:

lui_template = new_thread()

lui_template("Repeat this out: The best jokes are about 19th century scientists")

lui_template.thread_id

lui_template = new_thread() lui_template("Repeat this out: The best jokes are about 19th century scientists") lui_template.thread_id

🤖 LouieAI Response

Thread: D_SUNGJP8qPczYFtF6jl8M | Time: 2.9s

The best jokes are about 19th century scientists

Out[19]:

'D_SUNGJP8qPczYFtF6jl8M'

In [20]:

lui_joke = new_thread()

lui_joke(f"Read dthread {lui_template.thread_id}, and only after, make a joke about it")

lui_joke = new_thread() lui_joke(f"Read dthread {lui_template.thread_id}, and only after, make a joke about it")

🤖 LouieAI Response

Thread: D_h5umv0BvQ8GM5OsxtHTJ | Time: 9.8s

I need to read the specified dthread and then make a joke about its content.
NotebookAgent
read dthread D_SUNGJP8qPczYFtF6jl8MFinal Answer: Why did the 19th-century scientist bring a ladder to the bar? Because they heard the drinks were on the house, and they wanted to reach

[CallElement] read dthread D_SUNGJP8qPczYFtF6jl8M

| Markdown:





| Markdown: Repeat this out: The best jokes are about 19th century scientists


| Markdown: The best jokes are about 19th century scientists!



LouieAgent::runner
Input text:::Repeat this out: The best jokes are about 19th century scientists
Reading chat history. (is_prepared: True)
...
Louie output text: The best jokes are about 19th century scientists!

The notebook with the ID "D_SUNGJP8qPczYFtF6jl8M" contains a humorous statement emphasizing that the best jokes are about 19th-century scientists. This suggests a playful engagement with historical scientific figures, possibly highlighting their quirks or contributions in a light-hearted manner. The content reflects a blend of humor and historical interest, which could appeal to those who enjoy both science and comedy.

Out[20]:

🤖 LouieAI Response

I need to read the specified dthread and then make a joke about its content.
NotebookAgent
read dthread D_SUNGJP8qPczYFtF6jl8MFinal Answer: Why did the 19th-century scientist bring a ladder to the bar? Because they heard the drinks were on the house, and they wanted to reach

[CallElement] read dthread D_SUNGJP8qPczYFtF6jl8M

<notebook id="D_SUNGJP8qPczYFtF6jl8M">
<cells>
<cell id="B_w1xc1KB_" agent="LouieAgent" last_run_id="None">
<inputs>
<element type="TextElement"> | Markdown: </element>
</inputs>
<outputs/>
</cell>
<cell id="B_o1xgTGtG" agent="LouieAgent" last_run_id="R__vpblOybhMvl_vB49o_R">
<inputs>
<element type="TextElement"> | Markdown: Repeat this out: The best jokes are about 19th century scientists</element>
</inputs>
<outputs>
<element type="TextElement"> | Markdown: The best jokes are about 19th century scientists!</element>
</outputs>
<run id="R__vpblOybhMvl_vB49o_R" state="Done" run_type="LouieAgent">
<traces>
<trace level="Level.INFO">LouieAgent::runner</trace>
<trace level="Level.DEBUG">Input text:::Repeat this out: The best jokes are about 19th century scientists</trace>
<trace level="Level.DEBUG">Reading chat history. (is_prepared: True)</trace>
<trace level="Level.DEBUG">...</trace>
<trace level="Level.INFO">Louie output text: The best jokes are about 19th century scientists!</trace>
</traces>
</run>
</cell>
</cells>
</notebook>

The notebook with the ID "D_SUNGJP8qPczYFtF6jl8M" contains a humorous statement emphasizing that the best jokes are about 19th-century scientists. This suggests a playful engagement with historical scientific figures, possibly highlighting their quirks or contributions in a light-hearted manner. The content reflects a blend of humor and historical interest, which could appeal to those who enjoy both science and comedy.

---

✅ Session: Active | Thread ID: D_h5umv0BvQ8GM5OsxtHTJ | View Thread ↗ | Org: example-org

📚 History: 1 responses (access with lui[-1], lui[-2], etc.)

🔍 Traces: Disabled (use lui.traces = True to enable)

Latest Response:

• 3 text element(s) - access with lui.text or lui.texts

Quick Help (click to expand)

# Make a query

lui('Show me sales data from last week')


# Control visibility

lui('query', share_mode='Private')       # Default: only you

lui('query', share_mode='Organization')  # Share within org

lui('query', share_mode='Public')        # Share publicly


# Access results

df = lui.df          # Latest dataframe

text = lui.text      # Latest text response

all_dfs = lui.dfs    # All dataframes


# History

lui[-1].df           # Previous response's dataframe


# Traces (AI reasoning)

lui.traces = True    # Enable for session

lui('query', traces=True)  # Enable for one query

In [21]:

print(lui_joke.text)

print(lui_joke.text)

The notebook with the ID "D_SUNGJP8qPczYFtF6jl8M" contains a humorous statement emphasizing that the best jokes are about 19th-century scientists. This suggests a playful engagement with historical scientific figures, possibly highlighting their quirks or contributions in a light-hearted manner. The content reflects a blend of humor and historical interest, which could appeal to those who enjoy both science and comedy.

Next Steps¶

🎉 Congratulations! You've completed Part I of the Vibes Investigation series.

Continue your journey:

• Part II: Planning with Louie - Learn how to use Louie for strategic planning, roadmapping, and structured problem-solving

Other resources:

• Louie Documentation
• PyGraphistry Documentation