Chat using Functions with OpenAI (ChatGPT) API on New Column from Microsoft SQL Server API

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

Trigger workflow on

New Column from the Microsoft SQL Server API

Next, do this

Chat using Functions 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 Microsoft SQL Server 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 Column trigger
1. Connect your Microsoft SQL Server account
2. Configure Polling schedule
3. Select a Table
Configure the Chat using Functions action
1. Connect your OpenAI (ChatGPT) account
2. Configure alert
3. Select a Model
4. Configure Chat Input
5. Configure Functions
6. Optional- Configure Instructions
7. Optional- Select a Tool Choice
8. Optional- Configure Parallel Function Calling
9. Optional- Configure Previous Response ID
10. Optional- Select a Truncation
11. Optional- Select a Response Format
12. Optional- Configure Skip This Step
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 Column on Microsoft SQL Server

Description:Triggers when a new column is added to a table. [See the documentation](https://learn.microsoft.com/en-us/sql/relational-databases/system-catalog-views/sys-columns-transact-sql?view=sql-server-ver16)

Version:0.0.5

Key:microsoft_sql_server-new-column

View on GitHub

Microsoft SQL Server Overview#

With the Microsoft SQL Server API on Pipedream, you can automate your database operations and create powerful integrations. This API allows you to execute queries, manage databases, and trigger actions based on data changes. Implementing workflows that react to database events, manipulate data, or synchronize SQL Server data with other services, becomes effortless with Pipedream's serverless platform.

Trigger Code#

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

export default {
  ...common,
  key: "microsoft_sql_server-new-column",
  name: "New Column",
  description: "Triggers when a new column is added to a table. [See the documentation](https://learn.microsoft.com/en-us/sql/relational-databases/system-catalog-views/sys-columns-transact-sql?view=sql-server-ver16)",
  type: "source",
  version: "0.0.5",
  dedupe: "unique",
  props: {
    ...common.props,
    table: {
      propDefinition: [
        common.props.app,
        "table",
      ],
    },
  },
  methods: {
    ...common.methods,
    getCurrentColumns() {
      return this.db.get(constants.CURRENT_COLUMNS) || [];
    },
    setCurrentColumns(columns) {
      this.db.set(constants.CURRENT_COLUMNS, columns);
    },
    generateMeta(resource) {
      return {
        id: uuidv4(),
        summary: `New Column Added: ${resource.COLUMN_NAME}`,
        ts: Date.now(),
      };
    },
    async listResults() {
      const {
        app,
        table,
        getCurrentColumns,
      } = this;

      const columns = getCurrentColumns();
      const { recordset } = await app.listNewColumns({
        table,
        columns,
      });

      const newColumns = recordset.map(({ COLUMN_NAME: column }) => column);

      this.processResources(recordset);
      this.setCurrentColumns(newColumns.concat(columns));
    },
  },
};

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
Microsoft SQL Server	`app`	`app`	This component uses the Microsoft SQL Server app.
N/A	`db`	`$.service.db`	This component uses `$.service.db` to maintain state between executions.
Polling schedule	`timer`	`$.interface.timer`	How often to poll the API
Table	`table`	`string`	Select a value from the drop down menu.

Trigger Authentication#

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

Before you connect to your Microsoft SQL Server database from Pipedream, please make sure your database is either:

Accessible from the public internet (You may need to add a firewall rule on 0.0.0.0/0 on port 1433), or
Accessible from a static IP that you've configured from a VPC in Pipedream, and enabled the workflow to connect through that VPC

About Microsoft SQL Server#

Unleash the power of your data—on-premises, in the cloud, or at the edge—and reveal insights to transform your business.

Action#

Chat using Functions on OpenAI (ChatGPT)

Description:Chat with your models and allow them to invoke functions. Optionally, you can build and invoke workflows as functions. [See the documentation](https://platform.openai.com/docs/guides/function-calling)

Version:0.0.4

Key:openai-chat-using-functions

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";

export default {
  ...common,
  name: "Chat using Functions",
  version: "0.0.4",
  key: "openai-chat-using-functions",
  description: "Chat with your models and allow them to invoke functions. Optionally, you can build and invoke workflows as functions. [See the documentation](https://platform.openai.com/docs/guides/function-calling)",
  type: "action",
  props: {
    openai,
    alert: {
      type: "alert",
      alertType: "info",
      content: "Provide function names and parameters, and the model will either answer the question directly or decide to invoke one of the functions, returning a function call that adheres to your specified schema. Add a custom code step that includes all available functions which can be invoked based on the model's response - [you can even build an entire workflow as a function](https://pipedream.com/docs/workflows/building-workflows/code/nodejs/#invoke-another-workflow)! Once the appropriate function or workflow is executed, continue the overall execution or pass the result back to the model for further analysis. For more details, [see this guide](https://platform.openai.com/docs/guides/function-calling?api-mode=responses#overview) and this [walkthrough](https://pipedream.com/blog/introducing-enhanced-openai-chat-actions-for-pipedream/#using-pipedream-workflows-as-functions).",
    },
    modelId: {
      propDefinition: [
        openai,
        "chatCompletionModelId",
      ],
    },
    input: {
      type: "string",
      label: "Chat Input",
      description: "Text inputs to the model used to generate a response",
    },
    functions: {
      type: "string",
      label: "Functions",
      description: "A valid JSON array of functions using OpenAI's function schema definition. [See guide here](https://platform.openai.com/docs/guides/function-calling?api-mode=responses&example=search-knowledge-base#defining-functions).",
      default:
`[
  {
    "type": "function",
    "name": "your_function_name",
    "description": "Details on when and how to use the function",
    "strict": true,
    "parameters": {
      "type": "object",
      "properties": {
        "property_name": {
          "type": "property_type",
          "description": "A description for this property"
        },
        "another_property_name": {
          "type": "property_type",
          "description": "A description for this property"
        }
      },
      "required": [
        "list",
        "of",
        "required",
        "properties",
        "for",
        "this",
        "object"
      ],
      "additionalProperties": false
    }
  }
]`,
    },
    instructions: {
      type: "string",
      label: "Instructions",
      description: "Inserts a system (or developer) message as the first item in the model's context",
      optional: true,
    },
    toolChoice: {
      type: "string",
      label: "Tool Choice",
      description: "- **auto**: The model decides whether and how many functions to call.\n- **required**: The model must call one or more functions.\n- **function_name**: Enter a custom expression to force the model to call this specific function.",
      optional: true,
      default: "auto",
      options: [
        "auto",
        "required",
      ],
    },
    parallelToolCalls: {
      type: "boolean",
      label: "Parallel Function Calling",
      description: "Allow or prevent the model to call multiple functions in a single turn",
      optional: true,
      default: true,
    },
    previousResponseId: {
      type: "string",
      label: "Previous Response ID",
      description: "The unique ID of the previous response to the model. Use this to create multi-turn conversations",
      optional: true,
    },
    truncation: {
      type: "string",
      label: "Truncation",
      description: "Specifies the truncation mode for the response if it's larger than the context window size",
      optional: true,
      default: "auto",
      options: [
        "auto",
        "disabled",
      ],
    },
    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: [
        "text",
        "json_schema",
      ],
      default: "text",
      optional: true,
      reloadProps: true,
    },
    skipThisStep: {
      type: "boolean",
      label: "Skip This Step",
      description: "Pass in a boolean custom expression to skip this step's execution at runtime",
      optional: true,
      default: false,
    },
  },
  additionalProps() {
    const {
      modelId,
      responseFormat,
    } = this;
    const props = {};

    if (this.openai.isReasoningModel(modelId)) {
      props.reasoningEffort = {
        type: "string",
        label: "Reasoning Effort",
        description: "Constrains effort on reasoning for reasoning models",
        optional: true,
        options: [
          "low",
          "medium",
          "high",
        ],
      };

      // aparrently not supported yet as of 12/march/2025
      // props.generateSummary = {
      //   type: "string",
      //   label: "Generate Reasoning Summary",
      //   description: "A summary of the reasoning performed by the model",
      //   optional: true,
      //   options: [
      //     "concise",
      //     "detailed",
      //   ],
      // };
    }

    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. [Generate one here](https://platform.openai.com/docs/guides/structured-outputs/supported-schemas).",
      };
    }

    return props;
  },
  methods: {
    ...common.methods,
  },
  async run({ $ }) {
    if (this.skipThisStep) {
      $.export("$summary", "Step execution skipped");
      return;
    }

    const data = {
      model: this.modelId,
      input: this.input,
      instructions: this.instructions,
      previous_response_id: this.previousResponseId,
      truncation: this.truncation,
      parallel_tool_calls: this.parallelToolCalls,
      tools: [],
    };

    let functions = this.functions;
    if (typeof functions === "string") {
      try {
        functions = JSON.parse(functions);
      } catch (error) {
        throw new Error("Invalid JSON format in the provided Functions Schema");
      }
    }

    if (Array.isArray(functions)) {
      data.tools.push(...functions);
    } else {
      data.tools.push(functions);
    }

    if (this.toolChoice) {
      if (this.toolChoice === "auto" || this.toolChoice === "required") {
        data.tool_choice = this.toolChoice;
      } else {
        data.tool_choice = {
          type: "function",
          name: this.toolChoice,
        };
      }
    }

    if (this.openai.isReasoningModel(this.modelId) && this.reasoningEffort) {
      data.reasoning = {
        ...data.reasoning,
        effort: this.reasoningEffort,
      };
    }

    if (this.openai.isReasoningModel(this.modelId) && this.generateSummary) {
      data.reasoning = {
        ...data.reasoning,
        generate_summary: this.generateSummary,
      };
    }

    if (this.responseFormat === constants.CHAT_RESPONSE_FORMAT.JSON_SCHEMA.value) {
      try {
        data.text = {
          format: {
            type: this.responseFormat,
            ...JSON.parse(this.jsonSchema),
          },
        };
      } catch (error) {
        throw new Error("Invalid JSON format in the provided JSON Schema");
      }
    }

    const response = await this.openai.responses({
      $,
      data,
    });

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

    return 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.
Chat Input	`input`	`string`	Text inputs to the model used to generate a response
Functions	`functions`	`string`	A valid JSON array of functions using OpenAI's function schema definition. See guide here
Instructions	`instructions`	`string`	Inserts a system (or developer) message as the first item in the model's context
Tool Choice	`toolChoice`	`string`	Select a value from the drop down menu:`autorequired`
Parallel Function Calling	`parallelToolCalls`	`boolean`	Allow or prevent the model to call multiple functions in a single turn
Previous Response ID	`previousResponseId`	`string`	The unique ID of the previous response to the model. Use this to create multi-turn conversations
Truncation	`truncation`	`string`	Select a value from the drop down menu:`autodisabled`
Response Format	`responseFormat`	`string`	Select a value from the drop down menu:`textjson_schema`
Skip This Step	`skipThisStep`	`boolean`	Pass in a boolean custom expression to skip this step's execution at runtime

Chat using Functions with OpenAI (ChatGPT) API on New Column from Microsoft SQL Server API

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

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

Developers ♥ Pipedream

1-24of2,500+apps by most popular

1
-
24
of
2,500+
apps by most popular