Chat with Anthropic (Claude) API on New Run State Changed from OpenAI (ChatGPT) API

Pipedream makes it easy to connect APIs for Anthropic (Claude), OpenAI (ChatGPT) and 2,400+ other apps remarkably fast.

Trigger workflow on

New Run State Changed from the OpenAI (ChatGPT) API

Next, do this

Chat with the Anthropic (Claude) 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 OpenAI (ChatGPT) trigger and Anthropic (Claude) action. When you configure and deploy the workflow, it will run on Pipedream's servers 24x7 for free.

Select this integration
Configure the New Run State Changed trigger
1. Connect your OpenAI (ChatGPT) account
2. Configure timer
3. Configure Thread ID
Configure the Chat action
1. Connect your Anthropic (Claude) account
2. Select a Model
3. Configure User Message
4. Optional- Configure Prior Message History
5. Optional- Configure Temperature
6. Optional- Configure Top K
7. Optional- Configure Top P
8. Configure Maximum Tokens To Sample
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 Run State Changed on OpenAI (ChatGPT)

Description:Emit new event every time a run changes its status. [See the documentation](https://platform.openai.com/docs/api-reference/runs/listRuns)

Version:0.0.6

Key:openai-new-run-state-changed

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:

Trigger Code#

import common from "../common/common.mjs";
import sampleEmit from "./test-event.mjs";

export default {
  ...common,
  key: "openai-new-run-state-changed",
  name: "New Run State Changed",
  description: "Emit new event every time a run changes its status. [See the documentation](https://platform.openai.com/docs/api-reference/runs/listRuns)",
  version: "0.0.6",
  type: "source",
  dedupe: "unique",
  props: {
    ...common.props,
    threadId: {
      propDefinition: [
        common.props.openai,
        "threadId",
      ],
    },
  },
  methods: {
    ...common.methods,
    getStatusItems() {
      return this.db.get("statusItems");
    },
    setStatusItems(value) {
      this.db.set("statusItems", value);
    },
    async getData() {
      const {
        openai,
        threadId,
      } = this;

      const { data } = await openai.listRuns({
        threadId,
      });

      return data;
    },
    getMeta(item) {
      const ts = Date.now();
      return {
        id: `${item.id}-${ts}`,
        summary: `Run State Changed: ${item.id}`,
        ts,
      };
    },
    statusByItemId(itemId, statusItems) {
      const { [itemId]: status } = statusItems || {};
      return status;
    },
    buildStatusItems(data) {
      return data?.reduce((acc, item) => ({
        ...acc,
        [item.id]: item.status,
      }), {});
    },
    async getAndProcessItems(maxEvents) {
      const {
        getData,
        getStatusItems,
        setStatusItems,
        statusByItemId,
        buildStatusItems,
        getMeta,
        $emit: emit,
      } = this;

      const data = await getData();
      const statusItems = getStatusItems();

      Array.from(data)
        .reverse()
        .forEach((item, index) => {
          const statusChanged = statusByItemId(item.id, statusItems) !== item.status;
          if ((!maxEvents || index < maxEvents) && statusChanged) {
            emit(item, getMeta(item));
          }
        });

      setStatusItems(buildStatusItems(data));
    },
  },
  sampleEmit,
};

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
OpenAI (ChatGPT)	`openai`	`app`	This component uses the OpenAI (ChatGPT) app.
N/A	`db`	`$.service.db`	This component uses `$.service.db` to maintain state between executions.
	`timer`	`$.interface.timer`
Thread ID	`threadId`	`string`	The unique identifier for the thread. Example: `thread_abc123`. To locate the thread ID, make sure your OpenAI Threads setting (Settings -> Organization/Personal -> General -> Features and capabilities -> Threads) is set to "Visible to organization owners" or "Visible to everyone". You can then access the list of threads and click on individual threads to reveal their IDs

Trigger Authentication#

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

About OpenAI (ChatGPT)#

OpenAI is an AI research and deployment company with the mission to ensure that artificial general intelligence benefits all of humanity. They are the makers of popular models like ChatGPT, DALL-E, and Whisper.

Action#

Chat on Anthropic (Claude)

Description:The Chat API. [See the documentation](https://docs.anthropic.com/claude/reference/messages_post)

Version:0.0.9

Key:anthropic-chat

View on GitHub

Anthropic (Claude) Overview#

The Anthropic (Claude) API offers a conversational AI that can be leveraged for a variety of applications, ranging from answering questions to generating content or providing customer support. In Pipedream, you can create powerful workflows that utilize the capabilities of Claude to automate responses, analyze text, or enhance data with AI-generated insights. Pipedream's serverless platform simplifies integrating Claude into your processes with other apps and services for seamless automation.

Action Code#

import anthropic from "../../anthropic.app.mjs";
import constants from "../common/constants.mjs";

export default {
  name: "Chat",
  version: "0.0.9",
  key: "anthropic-chat",
  description: "The Chat API. [See the documentation](https://docs.anthropic.com/claude/reference/messages_post)",
  type: "action",
  props: {
    anthropic,
    model: {
      label: "Model",
      description: "Select the model to use for your query. Defaults to the `claude-3-opus-20240229` model, which Anthropic describes as the \"Most powerful model for highly complex tasks\".",
      type: "string",
      options: constants.MESSAGE_MODELS,
      default: constants.MESSAGE_MODELS[0],
    },
    userMessage: {
      label: "User Message",
      type: "string",
      description: "The user messages provide instructions to the assistant",
    },
    messages: {
      label: "Prior Message History",
      type: "string[]",
      description: "All relevant information must be supplied via the conversation. You can provide an array of messages from prior conversations here always beginning with the human message.",
      optional: true,
    },
    temperature: {
      label: "Temperature",
      description: "**Optional**. Amount of randomness injected into the response. Ranges from 0 to 1. Use temp closer to 0 for analytical / multiple choice, and temp closer to 1 for creative and generative tasks.",
      type: "string",
      optional: true,
    },
    topK: {
      label: "Top K",
      description: "Only sample from the top K options for each subsequent token. Used to remove `long tail` low probability responses.",
      type: "integer",
      optional: true,
    },
    topP: {
      label: "Top P",
      description: "Does nucleus sampling, in which we compute the cumulative distribution over all the options for each subsequent token in decreasing probability order and cut it off once it reaches a particular probability specified.",
      type: "string",
      optional: true,
    },
    maxTokensToSample: {
      label: "Maximum Tokens To Sample",
      description: "A maximum number of tokens to generate before stopping.",
      type: "integer",
    },
  },
  async run({ $ }) {
    const messages = [];

    const priorMessages = typeof this.messages === "string"
      ? JSON.parse(this.messages)
      : this.messages;

    if (priorMessages?.length) {
      let isUserMessage = true;

      for (const message of priorMessages) {
        messages.push({
          role: isUserMessage
            ? "user"
            : "assistant",
          content: message,
        });

        isUserMessage = !isUserMessage;
      }
    }

    messages.push({
      role: "user",
      content: this.userMessage,
    });

    const response = await this.anthropic.createMessage({
      $,
      data: {
        messages,
        model: this.model,
        max_tokens: this.maxTokensToSample,
        temperature: this.temperature
          ? parseFloat(this.temperature)
          : undefined,
        top_p: this.topP
          ? parseFloat(this.topP)
          : undefined,
        top_k: this.topK,
      },
    });

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

    const originalMessages = messages.map(({ content }) => content);
    return {
      original_messages: originalMessages,
      original_messages_with_assistant_response: originalMessages.concat(response.content[0].text),
      ...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
Anthropic (Claude)	`anthropic`	`app`	This component uses the Anthropic (Claude) app.
Model	`model`	`string`	Select a value from the drop down menu:`claude-3-5-sonnet-20241022claude-3-5-sonnet-20240620claude-3-opus-20240229claude-3-sonnet-20240229claude-3-haiku-20240307claude-2.1claude-2.0claude-instant-1.2`
User Message	`userMessage`	`string`	The user messages provide instructions to the assistant
Prior Message History	`messages`	`string[]`	All relevant information must be supplied via the conversation. You can provide an array of messages from prior conversations here always beginning with the human message.
Temperature	`temperature`	`string`	Optional. Amount of randomness injected into the response. Ranges from 0 to 1. Use temp closer to 0 for analytical / multiple choice, and temp closer to 1 for creative and generative tasks.
Top K	`topK`	`integer`	Only sample from the top K options for each subsequent token. Used to remove `long tail` low probability responses.
Top P	`topP`	`string`	Does nucleus sampling, in which we compute the cumulative distribution over all the options for each subsequent token in decreasing probability order and cut it off once it reaches a particular probability specified.
Maximum Tokens To Sample	`maxTokensToSample`	`integer`	A maximum number of tokens to generate before stopping.