Chat with OpenAI (ChatGPT) API on New Query Results from Snowflake API

Pipedream makes it easy to connect APIs for OpenAI (ChatGPT), Snowflake and 2,500+ other apps remarkably fast.

Trigger workflow on

New Query Results from the Snowflake API

Next, do this

Chat with the OpenAI (ChatGPT) API

No credit card required

▶

Watch us build a workflow

8 min

Watch now ➜

Trusted by 1,000,000+ developers from startups to Fortune 500 companies

Developers ♥ Pipedream

Getting Started#

This integration creates a workflow with a Snowflake trigger and OpenAI (ChatGPT) action. When you configure and deploy the workflow, it will run on Pipedream's servers 24x7 for free.

Select this integration
Configure the New Query Results trigger
1. Connect your Snowflake account
2. Configure timer
3. Configure SQL Query
4. Optional- Configure Primary key
5. Optional- Configure Emit individual events
Configure the Chat action
1. Connect your OpenAI (ChatGPT) account
2. Configure alert
3. Select a Model
4. Configure User Message
5. Optional- Configure Max Tokens
6. Optional- Configure Temperature
7. Optional- Configure Top P
8. Optional- Configure N
9. Optional- Configure Stop
10. Optional- Configure Presence Penalty
11. Optional- Configure Frequency Penalty
12. Optional- Configure User
13. Optional- Configure System Instructions
14. Optional- Configure Prior Message History
15. Optional- Configure Images
16. Optional- Configure Audio
17. Optional- Select a Response Format
18. Optional- Select one or more Tool Types
Deploy the workflow
Send a test event to validate your setup
Turn on the trigger

Details#

This integration uses pre-built, source-available components from Pipedream's GitHub repo. These components are developed by Pipedream and the community, and verified and maintained by Pipedream.

To contribute an update to an existing component or create a new component, create a PR on GitHub. If you're new to Pipedream component development, you can start with quickstarts for trigger span and action development, and then review the component API reference.

Trigger#

New Query Results on Snowflake

Description:Run a SQL query on a schedule, triggering a workflow for each row of results

Version:0.2.2

Key:snowflake-query-results

View on GitHub

Snowflake Overview#

Snowflake offers a cloud database and related tools to help developers create robust, secure, and scalable data warehouses. See Snowflake's Key Concepts & Architecture

Getting Started

1. Create a user, role and warehouse in Snowflake

Snowflake recommends you create a new user, role, and warehouse when you integrate a third-party tool like Pipedream. This way, you can control permissions via the user / role, and separate Pipedream compute and costs with the warehouse. You can do this directly in the Snowflake UI

We recommend you create a read-only account if you only need to query Snowflake. If you need to insert data into Snowflake, add permissions on the appropriate objects after you create your user.

2. Enter those details in Pipedream

Visit https://pipedream.com/accounts. Click the button to Connect an App. Enter the required Snowflake account data.

You'll only need to connect your account once in Pipedream. You can connect this account to multiple workflows to run queries against Snowflake, insert data, and more.

3. Build your first workflow

Visit https://pipedream.com/new to build your first workflow. Pipedream workflows let you connect Snowflake with 1,000+ other apps. You can trigger workflows on Snowflake queries, sending results to Slack, Google Sheets, or any app that exposes an API. Or you can accept data from another app, transform it with Python, Node.js, Go or Bash code, and insert it into Snowflake.

Learn more at Pipedream University

Trigger Code#

import common from "../common.mjs";
import { v4 as uuidv4 } from "uuid";

export default {
  ...common,
  type: "source",
  key: "snowflake-query-results",
  name: "New Query Results",
  // eslint-disable-next-line
  description: "Run a SQL query on a schedule, triggering a workflow for each row of results",
  version: "0.2.2",
  props: {
    ...common.props,
    sqlQuery: {
      type: "string",
      label: "SQL Query",
      description: "Run this query on a schedule, triggering the workflow for each row of results",
    },
    // Ordering props correctly
    timer: common.props.timer,
    dedupeKey: {
      type: "string",
      label: "Primary key",
      description: "The column in your query to use for de-duplication. Duplicate rows won't trigger your workflow",
      optional: true,
    },
    emitIndividualEvents: {
      propDefinition: [
        common.props.snowflake,
        "emitIndividualEvents",
      ],
    },
  },
  methods: {
    ...common.methods,
    generateMeta(data) {
      const {
        row: { [this.dedupeKey]: id = uuidv4() },
        timestamp: ts,
      } = data;
      const summary = `New event (ID: ${id})`;
      return {
        id,
        summary,
        ts,
      };
    },
    generateMetaForCollection(data) {
      const { timestamp: ts } = data;
      const id = uuidv4();
      const summary = `New event (ID: ${id})`;
      return {
        id,
        summary,
        ts,
      };
    },
    getStatement() {
      return {
        sqlText: this.sqlQuery,
      };
    },
  },
};

Trigger Configuration#

This component may be configured based on the props defined in the component code. Pipedream automatically prompts for input values in the UI and CLI.

Label	Prop	Type	Description
Snowflake	`snowflake`	`app`	This component uses the Snowflake app.
N/A	`db`	`$.service.db`	This component uses `$.service.db` to maintain state between executions.
	`timer`	`$.interface.timer`	Watch for changes on this schedule
SQL Query	`sqlQuery`	`string`	Run this query on a schedule, triggering the workflow for each row of results
Primary key	`dedupeKey`	`string`	The column in your query to use for de-duplication. Duplicate rows won't trigger your workflow
Emit individual events	`emitIndividualEvents`	`boolean`	Defaults to `true`, triggering workflows on each record in the result set. Set to `false` to emit records in batch (advanced)

Trigger Authentication#

Snowflake uses API keys for authentication. When you connect your Snowflake account, Pipedream securely stores the keys so you can easily authenticate to Snowflake APIs in both code and no-code steps.

We recommend you create a read-only account if you only need to query Snowflake. If you need to insert data into Snowflake, add permissions on the appropriate objects after you create your user.

About Snowflake#

A data warehouse built for the cloud

Action#

Chat on OpenAI (ChatGPT)

Description:The Chat API, using the `gpt-3.5-turbo` or `gpt-4` model. [See the documentation](https://platform.openai.com/docs/api-reference/chat)

Version:0.2.8

Key:openai-chat

View on GitHub

OpenAI (ChatGPT) Overview#

OpenAI provides a suite of powerful AI models through its API, enabling developers to integrate advanced natural language processing and generative capabilities into their applications. Here’s an overview of the services offered by OpenAI's API:

Use Python or Node.js code to make fully authenticated API requests with your OpenAI account:

Action Code#

import openai from "../../openai.app.mjs";
import common from "../common/common.mjs";
import constants from "../../common/constants.mjs";
import { ConfigurationError } from "@pipedream/platform";

export default {
  ...common,
  name: "Chat",
  version: "0.2.8",
  key: "openai-chat",
  description: "The Chat API, using the `gpt-3.5-turbo` or `gpt-4` model. [See the documentation](https://platform.openai.com/docs/api-reference/chat)",
  type: "action",
  props: {
    openai,
    alert: {
      type: "alert",
      alertType: "info",
      content: "Looking to chat with your tools? Check out our individual actions: [Chat using Web Search](https://pipedream.com/apps/openai/actions/chat-using-web-search), [Chat using File Search](https://pipedream.com/apps/openai/actions/chat-using-file-search), and [Chat using Functions](https://pipedream.com/apps/openai/actions/chat-using-functions).",
    },
    modelId: {
      propDefinition: [
        openai,
        "chatCompletionModelId",
      ],
    },
    userMessage: {
      label: "User Message",
      type: "string",
      description: "The user messages provide instructions to the assistant. They can be generated by the end users of an application, or set by a developer as an instruction.",
    },
    ...common.props,
    systemInstructions: {
      label: "System Instructions",
      type: "string",
      description: "The system message helps set the behavior of the assistant. For example: \"You are a helpful assistant.\" [See these docs](https://platform.openai.com/docs/guides/chat/instructing-chat-models) for tips on writing good instructions.",
      optional: true,
    },
    messages: {
      label: "Prior Message History",
      type: "string[]",
      description: "_Advanced_. Because [the models have no memory of past chat requests](https://platform.openai.com/docs/guides/chat/introduction), all relevant information must be supplied via the conversation. You can provide [an array of messages](https://platform.openai.com/docs/guides/chat/introduction) from prior conversations here. If this param is set, the action ignores the values passed to **System Instructions** and **Assistant Response**, appends the new **User Message** to the end of this array, and sends it to the API.",
      optional: true,
    },
    images: {
      label: "Images",
      type: "string[]",
      description: "Provide one or more images to [OpenAI's vision model](https://platform.openai.com/docs/guides/vision). Accepts URLs or base64 encoded strings. Compatible with the `gpt4-vision-preview` model",
      optional: true,
    },
    audio: {
      type: "string",
      label: "Audio",
      description: "Provide the file path to an audio file in the `/tmp` directory. For use with the `gpt-4o-audio-preview` model. Currently supports `wav` and `mp3` files.",
      optional: true,
    },
    responseFormat: {
      type: "string",
      label: "Response Format",
      description: "- **Text**: Returns unstructured text output.\n- **JSON Schema**: Enables you to define a [specific structure for the model's output using a JSON schema](https://platform.openai.com/docs/guides/structured-outputs?api-mode=responses).",
      options: Object.values(constants.CHAT_RESPONSE_FORMAT),
      default: constants.CHAT_RESPONSE_FORMAT.TEXT.value,
      optional: true,
      reloadProps: true,
    },
    toolTypes: {
      type: "string[]",
      label: "Tool Types",
      description: "The types of tools to enable on the assistant",
      options: constants.TOOL_TYPES.filter((toolType) => toolType === "function"),
      optional: true,
      reloadProps: true,
    },
  },
  additionalProps() {
    const {
      responseFormat,
      toolTypes,
      numberOfFunctions,
    } = this;
    const props = {};

    if (responseFormat === constants.CHAT_RESPONSE_FORMAT.JSON_SCHEMA.value) {
      props.jsonSchema = {
        type: "string",
        label: "JSON Schema",
        description: "Define the schema that the model's output must adhere to. [See the documentation here](https://platform.openai.com/docs/guides/structured-outputs/supported-schemas).",
      };
    }

    if (toolTypes?.includes("function")) {
      props.numberOfFunctions = {
        type: "integer",
        label: "Number of Functions",
        description: "The number of functions to define",
        optional: true,
        reloadProps: true,
        default: 1,
      };

      for (let i = 0; i < (numberOfFunctions || 1); i++) {
        props[`functionName_${i}`] = {
          type: "string",
          label: `Function Name ${i + 1}`,
          description: "The name of the function to be called. Must be a-z, A-Z, 0-9, or contain underscores and dashes, with a maximum length of 64.",
        };
        props[`functionDescription_${i}`] = {
          type: "string",
          label: `Function Description ${i + 1}`,
          description: "A description of what the function does, used by the model to choose when and how to call the function.",
          optional: true,
        };
        props[`functionParameters_${i}`] = {
          type: "object",
          label: `Function Parameters ${i + 1}`,
          description: "The parameters the functions accepts, described as a JSON Schema object. See the [guide](https://platform.openai.com/docs/guides/text-generation/function-calling) for examples, and the [JSON Schema reference](https://json-schema.org/understanding-json-schema/) for documentation about the format.",
          optional: true,
        };
      }
    }

    return props;
  },
  methods: {
    ...common.methods,
    _buildTools() {
      const tools = this.toolTypes?.filter((toolType) => toolType !== "function")?.map((toolType) => ({
        type: toolType,
      })) || [];
      if (this.toolTypes?.includes("function")) {
        const numberOfFunctions = this.numberOfFunctions || 1;
        for (let i = 0; i < numberOfFunctions; i++) {
          tools.push({
            type: "function",
            function: {
              name: this[`functionName_${i}`],
              description: this[`functionDescription_${i}`],
              parameters: this[`functionParameters_${i}`],
            },
          });
        }
      }
      return tools.length
        ? tools
        : undefined;
    },
  },
  async run({ $ }) {
    if (this.audio && !this.modelId.includes("gpt-4o-audio-preview")) {
      throw new ConfigurationError("Use of audio files requires using the `gpt-4o-audio-preview` model.");
    }

    const args = this._getChatArgs();

    const response = await this.openai.createChatCompletion({
      $,
      data: {
        ...args,
        tools: this._buildTools(),
      },
    });

    if (response) {
      $.export("$summary", `Successfully sent chat with id ${response.id}`);
    }

    const { messages } = args;
    return {
      original_messages: messages,
      original_messages_with_assistant_response: messages.concat(response.choices[0]?.message),
      ...response,
    };
  },
};

Action Configuration#

This component may be configured based on the props defined in the component code. Pipedream automatically prompts for input values in the UI.

Label	Prop	Type	Description
OpenAI (ChatGPT)	`openai`	`app`	This component uses the OpenAI (ChatGPT) app.
Model	`modelId`	`string`	Select a value from the drop down menu.
User Message	`userMessage`	`string`	The user messages provide instructions to the assistant. They can be generated by the end users of an application, or set by a developer as an instruction.
Max Tokens	`maxTokens`	`integer`	The maximum number of tokens to generate in the completion.
Temperature	`temperature`	`string`	Optional. What sampling temperature to use. Higher values means the model will take more risks. Try 0.9 for more creative applications, and 0 (argmax sampling) for ones with a well-defined answer.
Top P	`topP`	`string`	An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. So 0.1 means only the tokens comprising the top 10% probability mass are considered. We generally recommend altering this or `temperature` but not both.
N	`n`	`integer`	How many completions to generate for each prompt
Stop	`stop`	`string[]`	Up to 4 sequences where the API will stop generating further tokens. The returned text will not contain the stop sequence.
Presence Penalty	`presencePenalty`	`string`	Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the text so far, increasing the model's likelihood to talk about new topics.
Frequency Penalty	`frequencyPenalty`	`string`	Number between -2.0 and 2.0. Positive values penalize new tokens based on their existing frequency in the text so far, decreasing the model's likelihood to repeat the same line verbatim.
User	`user`	`string`	A unique identifier representing your end-user, which can help OpenAI to monitor and detect abuse. Learn more here
System Instructions	`systemInstructions`	`string`	The system message helps set the behavior of the assistant. For example: "You are a helpful assistant." See these docs for tips on writing good instructions.
Prior Message History	`messages`	`string[]`	Advanced. Because the models have no memory of past chat requests, all relevant information must be supplied via the conversation. You can provide an array of messages from prior conversations here. If this param is set, the action ignores the values passed to System Instructions and Assistant Response, appends the new User Message to the end of this array, and sends it to the API.
Images	`images`	`string[]`	Provide one or more images to OpenAI's vision model. Accepts URLs or base64 encoded strings. Compatible with the `gpt4-vision-preview` model
Audio	`audio`	`string`	Provide the file path to an audio file in the `/tmp` directory. For use with the `gpt-4o-audio-preview` model. Currently supports `wav` and `mp3` files.
Response Format	`responseFormat`	`string`	Select a value from the drop down menu:`{ "label": "Text", "value": "text" }{ "label": "JSON Object", "value": "json_object" }{ "label": "JSON Schema", "value": "json_schema" }`
Tool Types	`toolTypes`	`string[]`	Select a value from the drop down menu:`function`