What is Fack?

Fack answers frequently asked questions using Generative AI. Here is an example:

Use Cases

At Salesforce, we use fack for internal Q&A. We feed documents and summarized slack threads to fack.
In some channels, we see up to 40% of our support requests receive helpful answers from fack.

We are experiementing with other use cases like:

Improving incident resolution by quickly finding suggested solutions
Translating between query languages

Almost any RAG-based solution can be quickly implmented in fack.

Why is it Called Fack?

The term FAQ, or Frequently Asked Questions, is often pronounced fack.

Inspiration

Fack is built on the principles outlined by OpenAI for question/answer applications in documents like:

https://platform.openai.com/docs/tutorials/web-qa-embeddings
https://github.com/openai/openai-cookbook/blob/main/examples/Question_answering_using_embeddings.ipynb

Benefits of Fack

APIs

Reuse your documents across multiple systems. Need a Q/A interface? Done. Need to provide answers in a chat bot? Done.

Shared AI Knowledge

Working with LLMs is deceptively complex. Vector embeddings, token counting, effective prompting all require thought to make effective use of LLMs. By having a shared service, you can have one service to manage retrieval-augmented search and ask queries.

Deduplication

When managing thousands of documents, managing documents is difficult. Are there duplicates? What are the groups of documents?

Document Grouping

All documents are not equal. You have different groups within your company and organization. Different teams need to manage their documents. Visibilty may be an issue.

Principles

API first. The end user interactions will likely happen through other bots and applications. So, the service should be API first.
Multi-tenant. Different teams should be able to share the same app, without mixing their data.

Design

Data Flow

Why Search is Better than Fine-tuning

The OpenAI doc outlines the reasons why search/retrieval is typically better than fine-tuning.

"Fine-tuning is better suited to teaching specialized tasks or styles, and is less reliable for factual recall."
Search/retrival enables more reliable source citation. We can provide specific documents with urls/names to the GPT and require references to the documents used for answering the question. In fine-tuning, the original source is most likely lost or unavailable.

Developers

Requirements

Rails 7
Ruby 3+
Postgres with pg_vector support

Running Locally

Clone the repo

git clone https://github.com/salesforce/fack.git

Install dependencies

bundle

Database creation/migration

rails db:create
rails db:migrate
TEST_PASSWORD='<your_password>' rails db:seed

Set TEST_PASSWORD to an 8+ character string with a number, uppercase letter and special character. '2Testai!' as an example.

Create a .env file in the root directory. Provide these variables.

# LLM AUTH.  You need to provide an OpenAI Key or a Salesforce Einstein Org
## OpenAI key from https://platform.openai.com/account/api-keys
OPENAI_API_KEY=<openai token>

## OR Salesforce Einstein credentials from your Org
SALESFORCE_CONNECT_ORG_URL=
SALESFORCE_CONNECT_CLIENT_ID=
SALESFORCE_CONNECT_CLIENT_SECRET=
SALESFORCE_CONNECT_USERNAME=
SALESFORCE_CONNECT_PASSWORD=

# i.e. https://fack.yourdomain.com or http://localhost:3000 Used to generate URLs in the answers.
ROOT_URL=

## SAML/SSO Metadata URL (OPTIONAL)
SSO_METADATA_URL=

## Max number of document tokens to send to the GPT prompt. (OPTIONAL)
MAX_PROMPT_DOC_TOKENS=

## Max tokens to send in the prompt (OPTIONAL, DEFAULT 10,000)
EGPT_MAX_TOKENS=

## Which OpenAI model to use. (OPTIONAL)
EGPT_GEN_MODEL=

## Disable Password Login if you have SSO enabled.  (OPTIONAL)
DISABLE_PASSWORD_LOGIN=<true/false>

Open a new terminal and start the Background job for AI Calls

rake jobs:work

You should see: [Worker(host:host.something.com pid:89737)] Starting job worker

Start web server

rails s

Open http://localhost:3000 and login with admin@fack.com as the username and the TEST_PASSWORD you set earlier.

Testing

bundle exec rspec

The tests are located under the spec directory.

Usage

Create or Edit a Library

This page allows you to create or edit a library entry, providing key details like the name of the library, its source URL, and the owner of the library. Here’s how it works:

Library Name:
- Enter the name of the library in the provided text box. This could be the subject or title of the library, like "Data Science Resources."
Source URL:
- Provide the URL for the source of this library. This could be a link to the main website, document repository, or other related resources.
Owner Selection:
- Select the owner (administrator or user) who will manage this library. The dropdown list shows all available users, sorted by their email addresses. If no specific owner is needed, you can leave this field blank.

Ask a Question

This page allows you to ask a question and get an answer based on available libraries of knowledge. Here’s how it works:

Ask a Question:
- There is a text box where you can type your question. The placeholder text will give you examples of some topics the system can answer, based on the most frequently used libraries, such as specific subjects or categories.
Optional Library Selection:
- You can limit the answers to a specific library if you want. This means you can choose a particular topic or area for your question, and the system will only look for answers in that library. If you're unsure, you can leave it as "optional," and the system will check all available sources.
Submit Your Question:
- After typing your question, you can click the blue "Submit" button (which has an arrow icon) to send your question. The system will process your query and display the answer without refreshing the whole page, making it quicker and smoother.

This makes it easy to get answers quickly from a wide range of topics.

Assistants

The Assistant feature is designed to define Language Learning Model (LLM) assistants for specific topics. This form allows administrators or users to create, customize, and configure assistants according to their unique requirements. Below is an explanation of the fields and their purpose within the assistant creation form.

Fields

1. Status

Purpose: Defines the current status of the assistant.
Options:
- Ready
- Development

2. Owner (user_id)

Purpose: Assigns the owner of the assistant.
Type: Dropdown that displays users (emails) in the system.
Example: Select the appropriate user who will manage this assistant.

3. Name

Purpose: Specifies the name of the assistant.
Type: Text input.
Example: Provide a descriptive name for the assistant, such as "Healthcare Expert Assistant."

4. Quip URL

Purpose: Optional link to any relevant Quip document associated with the assistant.
Type: Text input.
Example: Add a Quip URL that holds documentation or collaboration content related to the assistant.

5. Confluence Spaces

Purpose: Lists the Confluence spaces that are associated with the assistant, if applicable.
Type: Text input.
Example: Enter space keys or names from Confluence for knowledge sharing related to the assistant.

6. Description

Purpose: Provides a detailed description of the assistant's purpose and functionality.
Type: Text area.
Example: A detailed explanation of what this assistant will help with, such as "This assistant answers common programming questions related to Ruby on Rails."

7. Libraries

Purpose: Lists the comma separated list of libraries the assistant will use.
Type: Text input.
Example: 2,3.

8. Input

Purpose: Describes the input that the assistant expects from the user.
Type: Text area.
Example: Describe what kind of input the assistant will work with, e.g., "natural language queries related to medical conditions."

9. Instructions

Purpose: Outlines any specific instructions on how the assistant will process the input.
Type: Text area.
Example: Define any special processing rules or behaviors, e.g., "translate medical terminology into layman's terms."

10. Output

Purpose: Describes the expected output or response that the assistant will provide.
Type: Text area.
Example: Specify the type of responses the assistant will generate, such as "detailed text responses to programming queries."

11. Context

Purpose: Defines the context or scope in which the assistant operates.
Type: Text area.
Example: "This assistant operates within the domain of healthcare research."

This structure provides flexibility in customizing assistants based on different user needs and topics, making it adaptable for various business or domain-specific requirements.

REST API

Questions

POST /api/v1/questions Create a new Question (Ask AI for answer)

Parameters

name type data type description

question required text The question to ask of the documentation

library_ids_included optional comma separated ids (reference to library) The libraries to limit the answers

Responses

http code content-type response

201 text/plain;charset=UTF-8 JSON Object

400 application/json {"code":"400","message":"Bad Request"}

Example cURL

curl -X POST -H "Authorization: Bearer <token>" -H "Content-Type: application/json" -d '{"question": { "question" : "how do i setup falcon?", library_ids_included: ["1"] }}' http://localhost:3000/api/v1/questions

{
  "id": 226,
  "question": "how do i setup falcon?",
  "status":"generating",
  "answer": "# ANSWER\nTo set up Falcon, you need to install the Falcon CLI. Here are the steps to install the Falcon CLI:\n\n1. For macOS users, install the Falcon CLI with brew:\n   ```\n   brew tap sfdc-falcon/cli git[@git.soma.salesforce.com:sfdc-falcon/>homebrew-cli.git](https://git.soma.salesforce.com/git.soma.salesforce.com:sfdc-falcon/homebrew-cli.git)\n   brew install falcon-cli\n   ```\n\n2.  For Linux users, install the Falcon CLI with `curl`:\n   ```\n   curl -sSL https://sfdc.co/get-falcon-cli | bash\n   ```\n\n3. Verify that you've successfully installed the CLI by logging in:\n   ```\n   falcon login\n   ```\n\nYou can find more information about setting up the Falcon CLI in the [Install the Falcon Command Line Interface (CLI)](https://git.soma.salesforce.com/tech-enablement/falcon-paved-path/blob/main/install-falcon-cli.md) document.\n\n# SOURCES\n- [Install the Falcon Command Line Interface (CLI)](https://git.soma.salesforce.com/tech-enablement/falcon-paved-path/blob/main/install-falcon-cli.md)",
  "created_at": "2023-11-03T17:28:43.625Z",
  "updated_at": "2023-11-03T17:28:43.625Z",
  "library_ids_included": ["1"],
  "url": "http://localhost:3000/questions/226.json"
}

GET /api/v1/questions/_id_ Retrieve Question

Parameters

name type data type description

Responses

name type data type description

question text The question to ask of the documentation

status pending, generating, generated, failed The status of the generated answer. Poll every 5 seconds until the status is generated or failed.

able_to_answer boolean Was the GPT able to generate an answer? (true/false)

url text URL to access the question

answer text Answer to the question

http code content-type response

200 text/plain;charset=UTF-8

400 application/json {"code":"400","message":"Bad Request"}

Example cURL

curl -X GET -H "Authorization: Bearer <token>"  http://localhost:3000/api/v1/questions/226

{
  "id": 226,
  "question": "how do i setup falcon?",
  "status":"generated",
  "answer": "# ANSWER\nTo set up Falcon, you need to install the Falcon CLI. Here are the steps to install the Falcon CLI:\n\n1. For macOS users, install the Falcon CLI with brew:\n   ```\n   brew tap sfdc-falcon/cli git[@git.soma.salesforce.com:sfdc-falcon/>homebrew-cli.git](https://git.soma.salesforce.com/git.soma.salesforce.com:sfdc-falcon/homebrew-cli.git)\n   brew install falcon-cli\n   ```\n\n2.  For Linux users, install the Falcon CLI with `curl`:\n   ```\n   curl -sSL https://sfdc.co/get-falcon-cli | bash\n   ```\n\n3. Verify that you've successfully installed the CLI by logging in:\n   ```\n   falcon login\n   ```\n\nYou can find more information about setting up the Falcon CLI in the [Install the Falcon Command Line Interface (CLI)](https://git.soma.salesforce.com/tech-enablement/falcon-paved-path/blob/main/install-falcon-cli.md) document.\n\n# SOURCES\n- [Install the Falcon Command Line Interface (CLI)](https://git.soma.salesforce.com/tech-enablement/falcon-paved-path/blob/main/install-falcon-cli.md)",
  "created_at": "2023-11-03T17:28:43.625Z",
  "updated_at": "2023-11-03T17:28:43.625Z",
  "url": "http://localhost:3000/questions/226.json"
}

GET /api/v1/questions List Questions

Parameters

name	type	data type	description
page	optional	integer	The page number to retrieve. Defaults to 1.

Responses

name	type	data type	description
questions		array	An array of question objects, each containing question details

Each object in the questions array includes:

name	data type	description
id	integer	The ID of the question
question	text	The content of the question
status	text	The status of the question (e.g., generated)
answer	text	The answer to the question
url	text	URL to access the question
created_at	datetime	The creation date and time of the question
updated_at	datetime	The last update date and time of the question

http code	content-type	response
`200`	`application/json`	JSON array of questions
`400`	`application/json`	`{"code":"400","message":"Bad Request"}`

Example cURL

curl -X GET -H "Authorization: Bearer <token>" "http://localhost:3000/api/v1/questions"

{
  "questions": [
    {
      "id": 226,
      "question": "how do i setup a new service?",
      "status": "generated",
      "answer": "# ANSWER\nTo set up ...",
      "created_at": "2023-11-03T17:28:43.625Z",
      "updated_at": "2023-11-03T17:28:43.625Z",
      "url": "http://localhost:3000/questions/226.json"
    },
    {
      "id": 227,
      "question": "how do i use gen ai?",
      "status": "generated",
      "answer": "# ANSWER\n...",
      "url": "http://localhost:3000/questions/227.json",
      "created_at": "2023-11-14T01:55:11.731Z",
      "updated_at": "2024-03-01T22:58:55.865Z"
    }
  ]
}

Documents

POST /api/v1/documents Create a new Document

Parameters

name type data type description

document required text The content of the document. 10,000 token limit.

title required text The title of the document

library_id required text The ID of the library to which this document will be added

external_id optional text A unique ID provided by the client. If a POST request includes the same external_id as an existing record, the record will be updated instead of created.

Example POST Data

Make sure you have the top level "document" attribute.

  "document": {
    "document": "Restart your computer to fix it.",
    "title": "How to fix a computer",
    "library_id": 23,
   }

Responses

http code content-type response

201 text/plain;charset=UTF-8 Document created successfully

400 application/json {"code":"400","message":"Bad Request"}

Example cURL

 curl -X POST -H "Content-Type: application/json" -H "Authorization: Bearer <token>" -d '{"document": {"document":"Document Content", "library_id":"your_library_id", "external_id":"optional_unique_id"}}' http://localhost:3000/api/v1/documents

GET /api/v1/documents/_id_ Retrieve Document

Parameters

name type data type description

Responses

name	type	data type	description
document		object	A document objects

The document object includes:

name	data type	description
id	integer	The ID of the document
document	text	The content of the document
title	text	The title of the document
url	text	URL to access the document
created_at	datetime	The creation date and time of the document
updated_at	datetime	The last update date and time of the document

http code content-type response

200 text/plain;charset=UTF-8

400 application/json {"code":"400","message":"Bad Request"}

Example cURL

 curl -X GET -H "Authorization: Bearer <token>" http://localhost:3000/api/v1/documents/<id>

{
 "id": 1,
 "document": "# QUESTION\nHow do I use gen ai?\n\n# ANSWER\n...",
 "title": "How do I use GenAI?",
 "url": "http://localhost:3000/documents/1",
 "length": 97,
 "created_at": "2023-11-14T01:55:11.731Z",
 "updated_at": "2024-03-01T22:58:55.865Z"
}

GET /api/v1/documents List Documents

Parameters

name	type	data type	description
page	optional	integer	The page number to retrieve. Defaults to 1.

Responses

name	type	data type	description
documents		array	An array of document objects, each containing document details

Each object in the documents array includes:

name	data type	description
id	integer	The ID of the document
document	text	The content of the document
title	text	The title of the document
url	text	URL to access the document
created_at	datetime	The creation date and time of the document
updated_at	datetime	The last update date and time of the document

http code	content-type	response
`200`	`application/json`	JSON array of documents
`400`	`application/json`	`{"code":"400","message":"Bad Request"}`

Example cURL

curl -X GET -H "Authorization: Bearer <token>" "http://localhost:3000/api/v1/documents?page=1"

{
 "documents": [
   {
     "id": 1,
     "document": "# QUESTION\nHow do I use gen ai?\n\n# ANSWER\nThere is no direct answer provided in the conversation.",
     "url": "http://localhost:3000/documents/1",
     "created_at": "2023-11-14T01:55:11.731Z",
     "updated_at": "2024-03-01T22:58:55.865Z"
   },
   {
     "id": 2,
     "document": "# QUESTION\nHow do I integrate API?\n\n# ANSWER\nTo integrate an API, first identify the API you need to integrate with...",
     "url": "http://localhost:3000/documents/2",
     "created_at": "2023-12-14T02:55:11.731Z",
     "updated_at": "2024-01-02T23:58:55.865Z"
   }
   ...
 ]
}

Libraries

POST /api/v1/libraries Create a new Library

Parameters

name type data type description

name required text The name of the library

Responses

http code content-type response

201 text/plain;charset=UTF-8 Library created successfully

400 application/json {"code":"400","message":"Bad Request"}

Example cURL

curl -X POST -H "Content-Type: application/json" -H "Authorization: Bearer <token>" -d '{"name":"Library Name"}' http://localhost:3000/api/v1/libraries

GET /api/v1/library/_id_ Retrieve Library

Parameters

name type data type description

Responses

http code content-type response

200 text/plain;charset=UTF-8

400 application/json {"code":"400","message":"Bad Request"}

Example cURL

curl -X GET -H "Authorization: Bearer <token>" http://localhost:3000/api/v1/libraries/<id>

{
 "id": 1,
 "name": "My Docs",
 "created_at": "2023-11-15T20:17:25.665Z",
 "updated_at": "2023-12-01T19:59:44.618Z",
 "url": "http://localhost:3000/libraries/1"
}

GET /api/v1/libraries List Libraries

Parameters

name	type	data type	description
page	optional	integer	The page number to retrieve. Defaults to 1.

Responses

name	type	data type	description
libraries		array	An array of library objects, each containing library details

Each object in the documents array includes:

name	data type	description
id	integer	The ID of the library
name	text	Name of the library
created_at	datetime	The creation date and time of the library
updated_at	datetime	The last update date and time of the library

http code	content-type	response
`200`	`application/json`	JSON array of libraries
`400`	`application/json`	`{"code":"400","message":"Bad Request"}`

Example cURL

curl -X GET -H "Authorization: Bearer <token>" "http://localhost:3000/api/v1/libraries"

{
 "libraries": [
   {
     "id": 1,
     "name": "My Library",
     "created_at": "2023-11-14T01:55:11.731Z",
     "updated_at": "2024-03-01T22:58:55.865Z"
   },
   {
     "id": 2,
     "name": "My Second Library",
     "created_at": "2023-01-14T01:55:11.731Z",
     "updated_at": "2023-03-01T22:58:55.865Z"
   },
   ...
 ]
}

WebSocket API

The MessagesChannel WebSocket API allows authenticated users to subscribe to a real-time messaging stream. Users must provide an authentication token to connect and stream messages from the channel.

WebSocket URL

To connect to the WebSocket API, use the following URL:

ws://yourserver.com/cable

Replace yourserver.com with the appropriate domain or IP where your Rails server is hosted.

Connection Steps

1. Establish the WebSocket Connection

To initiate a WebSocket connection, you must connect to the WebSocket endpoint and send a subscription request to the MessagesChannel with a valid token.

2. Using Native WebSocket API in JavaScript

If you want to use the native WebSocket API, here’s how you can establish a connection and subscribe to the MessagesChannel:

const apiToken = 'your_api_token_here';

// Establish WebSocket connection
const ws = new WebSocket('ws://yourserver.com/cable');

ws.onopen = function() {
  console.log('Connected to WebSocket!');

  // Send a subscription request to the MessagesChannel with the token
  const subscriptionMessage = {
    command: 'subscribe',
    identifier: JSON.stringify({ channel: 'MessagesChannel', token: apiToken })
  };

  ws.send(JSON.stringify(subscriptionMessage));
};

ws.onmessage = function(event) {
  const data = JSON.parse(event.data);

  // Ignore ping messages
  if (data.type === 'ping') return;

  // Handle messages from the server
  if (data.message) {
    console.log('Received message:', data.message);
  }
};

ws.onerror = function(error) {
  console.error('WebSocket error:', error);
};

ws.onclose = function() {
  console.log('Disconnected from WebSocket');
};

3. Handling Subscription and Disconnection

When subscribing to the MessagesChannel, messages from the server will follow the standard ActionCable format. Incoming messages might look like:

{
  "identifier": "{\"channel\":\"MessagesChannel\"}",
  "message": {
    "message": "Hello, World!"
  }
}

Ensure that your client is prepared to handle incoming data and filter out any ping or control messages.

Guides

Setup SSO

Get the SSO Metadata URL from your SSO provider. i.e. https://company.okta.com/app/xyz/sso/saml/metadata
Set the SSO_METADATA_URL to the url from previous step in your .env file or environment.
Restart your app.
Optionally, disable username/password login with DISABLE_PASSWORD_LOGIN=true environment variable.

Importing Your Documents (DRAFT)

1. Choose or Create a Library

The library is a collection of similar documents. For example:

Infrastructure
Dev Docs
Onboarding

The library allows document owners to keep their documents separate from each other. It also enables more selective question/answers.

Existing Library

Go to /libraries in the UI
Locate the library you want.
Note the id in the URL.

Creating a Library

Go to /libraries in the UI
Click "New Library" button.
Provide a name and save. Get the id of the library from the URL.

2. Get an API Token

Open /api_tokens in the ui.
Create a token.

3. Locate the Directory with your Documents

You can clone your doc repo or have an directory anywhere on your computer.

4. Set the variables in .env

IMPORT_API_TOKEN=<your token>
ROOT_URL=<root url>  # "http://localhost:3000" if you are testing locally.

5. Install jq

jq is a dependency.

See the download page for instructions.

6. Run the import script

./import.sh -l <library_id> -d <path_to_your_docs>

7. Verify and Test

Go to /libraries and click on the library you used.
See if the documents from your import are listed.

name	type	data type	description
question	required	text	The question to ask of the documentation
library_ids_included	optional	comma separated ids (reference to library)	The libraries to limit the answers

http code	content-type	response
`201`	`text/plain;charset=UTF-8`	JSON Object
`400`	`application/json`	`{"code":"400","message":"Bad Request"}`

http code	content-type	response
`201`	`text/plain;charset=UTF-8`	`Document created successfully`
`400`	`application/json`	`{"code":"400","message":"Bad Request"}`

http code	content-type	response
`201`	`text/plain;charset=UTF-8`	`Library created successfully`
`400`	`application/json`	`{"code":"400","message":"Bad Request"}`

Files

README.md

Latest commit

History

README.md

File metadata and controls

What is Fack?

Use Cases

Why is it Called Fack?

Inspiration

Benefits of Fack

APIs

Shared AI Knowledge

Deduplication

Document Grouping

Principles

Design

Data Flow

Why Search is Better than Fine-tuning

Developers

Requirements

Running Locally

Testing

Usage

Create or Edit a Library

Ask a Question

Assistants

Fields

1. Status

2. Owner (user_id)

3. Name

4. Quip URL

5. Confluence Spaces

6. Description

7. Libraries

8. Input

9. Instructions

10. Output

11. Context

REST API

Questions

Parameters

Responses

Example cURL

Parameters

Responses

Example cURL

Parameters

Responses

Example cURL

Documents

Parameters

Example POST Data

Responses

Example cURL

Parameters

Responses

Example cURL

Parameters

Responses

Example cURL

Libraries

Parameters

Responses

Example cURL

Parameters

Responses

Example cURL

Parameters

Responses

Example cURL

WebSocket API

WebSocket URL

Connection Steps

1. Establish the WebSocket Connection

2. Using Native WebSocket API in JavaScript

3. Handling Subscription and Disconnection

Guides

Setup SSO

Importing Your Documents (DRAFT)