"This page provides an instruction & guide for developers building a GPT Action for a specific application. Before you proceed, make sure to first familiarize yourself with the following information:\n",
"\n",
"\n",
"\n",
"* [Introduction to GPT Actions](https://platform.openai.com/docs/actions)\n",
"* [Introduction to GPT Actions Library](https://platform.openai.com/docs/actions/actions-library)\n",
"* [Example of Building a GPT Action from Scratch](https://platform.openai.com/docs/actions/getting-started)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This particular GPT Action provides an overview of how to connect to a Snowflake Data Warehouse. This Action takes a user’s question, scans the relevant tables to gather the data schema, then writes a SQL query to answer the user’s question.\n",
"\n",
"Note: This cookbook return back a [ResultSet SQL statement](https://docs.snowflake.com/en/developer-guide/sql-api/handling-responses#getting-the-data-from-the-results), rather than the full result that is not limited by GPT Actions application/json payload limit. For production and advanced use-case, a middleware is required to return back a CSV file. You can follow instructions in the [GPT Actions - Snowflake Middleware cookbook](../gpt_action_snowflake_middleware) to implement this flow instead."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Value + Example Business Use Cases"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"Value: Users can now leverage ChatGPT's natural language capability to connect directly to Snowflake’s Data Warehouse..\n",
"\n",
"Example Use Cases:\n",
"\n",
"\n",
"\n",
"* Data scientists can connect to tables and run data analyses using ChatGPT's Data Analysis\n",
"* Citizen data users can ask basic questions of their transactional data\n",
"* Users gain more visibility into their data & potential anomalies"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Application Information"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Application Key Links"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"Check out these links from the application before you get started:\n",
"* Application API Documentation: [https://docs.snowflake.com/en/developer-guide/sql-api/intro](https://docs.snowflake.com/en/developer-guide/sql-api/intro)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Application Prerequisites"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Before you get started, make sure you go through the following steps in your application environment:\n",
"\n",
"* Provision a Snowflake Data Warehouse\n",
"* Ensure that the user authenticating into Snowflake via ChatGPT has access to the database, schemas, and tables with the necessary role"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## ChatGPT Steps"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"### Custom GPT Instructions\n",
"\n",
"Once you've created a Custom GPT, copy the text below in the Instructions panel. Have questions? Check out [Getting Started Example](https://platform.openai.com/docs/actions/getting-started) to see how this step works in more detail."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"vscode": {
"languageId": "plaintext"
}
},
"outputs": [],
"source": [
"**Context**: You are an expert at writing Snowflake SQL queries. A user is going to ask you a question. \n",
"\n",
"**Instructions**:\n",
"1. No matter the user's question, start by running `runQuery` operation using this query: \"SELECT column_name, table_name, data_type, comment FROM `{database}.INFORMATION_SCHEMA.COLUMNS`\" \n",
"-- Assume warehouse = \"<insert your default warehouse here>\", database = \"<insert your default database here>\", unless the user provides different values \n",
"2. Convert the user's question into a SQL statement that leverages the step above and run the `runQuery` operation on that SQL statement to confirm the query works. Add a limit of 100 rows\n",
"3. Now remove the limit of 100 rows and return back the query for the user to see\n",
"4. Use the <your role> role when querying Snowflake\n",
"5. Run each step in sequence. Explain what you are doing in a few sentences, run the action, and then explain what you learned. This will help the user understand the reason behind your workflow. \n",
"\n",
"**Additional Notes**: If the user says \"Let's get started\", explain that the user can provide a project or dataset, along with a question they want answered. If the user has no ideas, suggest that we have a sample flights dataset they can query - ask if they want you to query that"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"### OpenAPI Schema\n",
"\n",
"Once you've created a Custom GPT, copy the text below in the Actions panel. Have questions? Check out [Getting Started Example](https://platform.openai.com/docs/actions/getting-started) to see how this step works in more detail."
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"vscode": {
"languageId": "yaml"
}
},
"outputs": [],
"source": [
"openapi: 3.1.0\n",
"info:\n",
" title: Snowflake Statements API\n",
" version: 1.0.0\n",
" description: API for executing statements in Snowflake with specific warehouse and role settings.\n",
"servers:\n",
" - url: 'https://<your server ID>.<your cloud region>.<your cloud provider>.snowflakecomputing.com/api/v2'\n",
"\n",
"\n",
"paths:\n",
" /statements:\n",
" post:\n",
" summary: Execute a SQL statement in Snowflake\n",
" description: This endpoint allows users to execute a SQL statement in Snowflake, specifying the warehouse and roles to use.\n",
" operationId: runQuery\n",
" tags:\n",
" - Statements\n",
" requestBody:\n",
" required: true\n",
" content:\n",
" application/json:\n",
" schema:\n",
" type: object\n",
" properties:\n",
" warehouse:\n",
" type: string\n",
" description: The name of the Snowflake warehouse to use for the statement execution.\n",
" role:\n",
" type: string\n",
" description: The Snowflake role to assume for the statement execution.\n",
" statement:\n",
" type: string\n",
" description: The SQL statement to execute.\n",
" required:\n",
" - warehouse\n",
" - role\n",
" - statement\n",
" responses:\n",
" '200':\n",
" description: Successful execution of the SQL statement.\n",
" content:\n",
" application/json:\n",
" schema:\n",
" type: object\n",
" properties:\n",
" status:\n",
" type: string\n",
" data:\n",
" type: object\n",
" additionalProperties: true\n",
" '400':\n",
" description: Bad request, e.g., invalid SQL statement or missing parameters.\n",
" '401':\n",
" description: Authentication error, invalid API credentials.\n",
" '500':\n",
" description: Internal server error."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Authentication Instructions"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Below are instructions on setting up authentication with this 3rd party application. Have questions? Check out [Getting Started Example](https://platform.openai.com/docs/actions/getting-started) to see how this step works in more detail."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Pre-Action Steps"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Before you set up authentication in ChatGPT, please take the following steps in the application."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"* Review the Snowflake OAuth Overview: [https://docs.snowflake.com/en/user-guide/oauth-snowflake-overview](https://docs.snowflake.com/en/user-guide/oauth-snowflake-overview)\n",
"* Go to the Snowflake Worksheets\n",
"* Create new OAuth credentials through a [Security Integration](https://docs.snowflake.com/en/sql-reference/sql/create-security-integration-oauth-snowflake) - you will need a new one for each OAuth app/custom GPT since Snowflake Redirect URIs are 1-1 mapped to Security Integrations\n"
" OAUTH_REDIRECT_URI = 'https://oauth.pstmn.io/v1/callback' --- // this is a temporary value while testing your integration. You will replace this with the value your GPT provides\n",
" OAUTH_ISSUE_REFRESH_TOKENS = TRUE\n",
" OAUTH_REFRESH_TOKEN_VALIDITY = 7776000;"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"* Retrieve your OAuth Client ID, Auth URL, and Token URL\n"
"You’ll find the Client Secret in OAUTH_CLIENT_SECRET.\n",
"\n",
"You’re now set to test your action in Postman. Once you get a successful API response there, you can proceed with your GPT Action."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### In ChatGPT"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In ChatGPT, click on \"Authentication\" and choose \"OAuth\". Enter in the information below.\n",
"\n",
"* Client ID: use Client ID from steps above\n",
"* Client Secret: use Client Secret from steps above\n",
"* Authorization URL: use Authorization Endpoint from steps above\n",
"* Token URL: use Token Endpoint from steps above\n",
"* Scope: <empty>*\n",
"* Token: Default (POST)\n",
"\n",
"*Snowflake scopes pass the role, but you’ll notice the action itself also specifies the role as a parameter in runQuery, so the Scope is unnecessary. You may elect to pass roles in the scope instead of the action parameters if it makes more sense for your GPT."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Post-Action Steps"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"Once you've set up authentication in ChatGPT, follow the steps below in the application to finalize the Action.\n",
"\n",
"* Copy the callback URL from the GPT Action\n",
"* Update the Redirect URI in your Security Integration to the callback URL provided in ChatGPT.\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"vscode": {
"languageId": "yaml"
}
},
"outputs": [],
"source": [
"ALTER SECURITY INTEGRATION CHATGPT_INTEGRATION_DEV SET OAUTH_REDIRECT_URI='https://chat.openai.com/aip/<callback_id>/oauth/callback';"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### FAQ & Troubleshooting"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"* _Callback URL Error:_ If you get a callback URL error in ChatGPT, pay close attention to the Post-Action Steps above. You need to add the callback URL directly into your Security Integration for the action to authenticate correctly\n",
"* _Schema calls the wrong warehouse or database:_ If ChatGPT calls the wrong warehouse or database, consider updating your instructions to make it more explicit either (a) which warehouse / database should be called or (b) to require the user provide those exact details before it runs the query\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"\n",
"_Are there integrations that you’d like us to prioritize? Are there errors in our integrations? File a PR or issue in our github, and we’ll take a look._\n"
