Choose Best Completion with Cohere API on New Scheduled Tasks from Pipedream API

Pipedream makes it easy to connect APIs for Cohere, Pipedream and 2,400+ other apps remarkably fast.

Trigger workflow on

New Scheduled Tasks from the Pipedream API

Next, do this

Choose Best Completion with the Cohere 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 Pipedream trigger and Cohere action. When you configure and deploy the workflow, it will run on Pipedream's servers 24x7 for free.

Select this integration
Configure the New Scheduled Tasks trigger
1. Connect your Pipedream account
2. Optional- Configure Secret
Configure the Choose Best Completion action
1. Connect your Cohere account
2. Configure Prompt
3. Optional- Select a Model
4. Optional- Configure Number of generations
5. Optional- Configure Max Tokens
6. Optional- Configure Preset
7. Optional- Configure Temperature
8. Optional- Configure K
9. Optional- Configure P
10. Optional- Configure Frequency Penalty
11. Optional- Configure End Sequences
12. Optional- Configure Stop Sequences
13. Optional- Select a Return Likelihoods
14. Optional- Configure Logit Bias
15. Optional- Select a Truncate
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 Scheduled Tasks on Pipedream

Description:Exposes an HTTP API for scheduling messages to be emitted at a future time

Version:0.3.1

Key:pipedream-new-scheduled-tasks

View on GitHub

Pipedream Overview#

Pipedream is an API that allows you to build applications that can connect to
various data sources and processes them in real-time. You can use Pipedream to
create applications that can perform ETL (Extract, Transform, and Load) tasks,
as well as to create data-driven workflows.

Some examples of applications you can build using the Pipedream API include:

An application that can extract data from a database, transform it, and then
load it into another database.
An application that can monitor a data source for changes, and then trigger a
workflow in response to those changes.
An application that can poll an API for new data, and then process that data
in real-time.

Trigger Code#

import pipedream from "../../pipedream.app.mjs";
import sampleEmit from "./test-event.mjs";
import { uuid } from "uuidv4";

export default {
  key: "pipedream-new-scheduled-tasks",
  name: "New Scheduled Tasks",
  type: "source",
  description:
    "Exposes an HTTP API for scheduling messages to be emitted at a future time",
  version: "0.3.1",
  dedupe: "unique", // Dedupe on a UUID generated for every scheduled task
  props: {
    pipedream,
    secret: {
      type: "string",
      secret: true,
      label: "Secret",
      optional: true,
      description:
        "**Optional but recommended**: if you enter a secret here, you must pass this value in the `x-pd-secret` HTTP header when making requests",
    },
    http: {
      label: "Endpoint",
      description: "The endpoint where you'll send task scheduler requests",
      type: "$.interface.http",
      customResponse: true,
    },
    db: "$.service.db",
  },
  methods: {
    // To schedule future emits, we emit to the selfChannel of the component
    selfChannel() {
      return "self";
    },
    // Queue for future emits that haven't yet been delivered
    queuedEventsChannel() {
      return "$in";
    },
    httpRespond({
      status, body,
    }) {
      this.http.respond({
        headers: {
          "content-type": "application/json",
        },
        status,
        body,
      });
    },
    async selfSubscribe() {
      // Subscribe the component to itself. We do this here because even in
      // the activate hook, the component isn't available to take subscriptions.
      // Scheduled tasks are sent to the self channel, which emits the message at
      // the specified delivery_ts to this component.
      const isSubscribedToSelf = this.db.get("isSubscribedToSelf");
      if (!isSubscribedToSelf) {
        const componentId = process.env.PD_COMPONENT;
        const selfChannel = this.selfChannel();
        console.log(`Subscribing to ${selfChannel} channel for event source`);
        console.log(
          await this.pipedream.subscribe(componentId, componentId, selfChannel),
        );
        this.db.set("isSubscribedToSelf", true);
      }
    },
    validateEventBody(event, operation) {
      const errors = [];

      // Secrets are optional, so we first check if the user configured
      // a secret, then check its value against the prop (validation below)
      if (this.secret && event.headers["x-pd-secret"] !== this.secret) {
        errors.push(
          "Secret on incoming request doesn't match the configured secret",
        );
      }

      if (operation === "schedule") {
        const {
          timestamp,
          message,
        } = event.body;
        // timestamp should be an ISO 8601 string. Parse and check for validity below.
        const epoch = Date.parse(timestamp);

        if (!timestamp) {
          errors.push(
            "No timestamp included in payload. Please provide an ISO8601 timestamp in the 'timestamp' field",
          );
        }
        if (timestamp && !epoch) {
          errors.push("Timestamp isn't a valid ISO 8601 string");
        }
        if (!message) {
          errors.push("No message passed in payload");
        }
      }

      return errors;
    },
    scheduleTask(event) {
      const errors = this.validateEventBody(event, "schedule");
      let status, body;

      if (errors.length) {
        console.log(errors);
        status = 400;
        body = {
          errors,
        };
      } else {
        const id = this.emitScheduleEvent(event.body, event.body.timestamp);
        status = 200;
        body = {
          msg: "Successfully scheduled task",
          id,
        };
      }

      this.httpRespond({
        status,
        body,
      });
    },
    emitScheduleEvent(event, timestamp) {
      const selfChannel = this.selfChannel();
      const epoch = Date.parse(timestamp);
      const $id = uuid();

      console.log(`Scheduled event to emit on: ${new Date(epoch)}`);

      this.$emit(
        {
          ...event,
          $channel: selfChannel,
          $id,
        },
        {
          name: selfChannel,
          id: $id,
          delivery_ts: epoch,
        },
      );

      return $id;
    },
    async cancelTask(event) {
      const errors = this.validateEventBody(event, "cancel");
      let status, msg;

      if (errors.length) {
        console.log(errors);
        status = 400;
        msg = "Secret on incoming request doesn't match the configured secret";
      } else {
        try {
          const id = event.body.id;
          const isCanceled = await this.deleteEvent(event);
          if (isCanceled) {
            status = 200;
            msg = `Cancelled scheduled task for event ${id}`;
          } else {
            status = 404;
            msg = `No event with ${id} found`;
          }
        } catch (error) {
          console.log(error);
          status = 500;
          msg = "Failed to schedule task. Please see the logs";
        }
      }

      this.httpRespond({
        status,
        body: {
          msg,
        },
      });
    },
    async deleteEvent(event) {
      const componentId = process.env.PD_COMPONENT;
      const inChannel = this.queuedEventsChannel();

      // The user must pass a scheduled event UUID they'd like to cancel
      // We lookup the event by ID and delete it
      const { id } = event.body;

      // List events in the $in channel - the queue of scheduled events, to be emitted in the future
      const events = await this.pipedream.listEvents(
        componentId,
        inChannel,
      );
      console.log("Events: ", events);

      // Find the event in the list by id
      const eventToCancel = events.data.find((e) => {
        const { metadata } = e;
        return metadata.id === id;
      });

      console.log("Event to cancel: ", eventToCancel);

      if (!eventToCancel) {
        console.log(`No event with ${id} found`);
        return false;
      }

      // Delete the event
      await this.pipedream.deleteEvent(
        componentId,
        eventToCancel.id,
        inChannel,
      );
      return true;
    },
    emitEvent(event, summary) {
      // Delete the channel name and id from the incoming event, which were used only as metadata
      const id = event.$id;
      delete event.$channel;
      delete event.$id;

      this.$emit(event, {
        summary: summary ?? JSON.stringify(event),
        id,
        ts: +new Date(),
      });
    },
  },
  async run(event) {
    await this.selfSubscribe();

    const { path } = event;
    if (path === "/schedule") {
      this.scheduleTask(event);
    } else if (path === "/cancel") {
      await this.cancelTask(event);
    } else if (event.$channel === this.selfChannel()) {
      this.emitEvent(event);
    }
  },
  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
Pipedream	`pipedream`	`app`	This component uses the Pipedream app.
Secret	`secret`	`string`	Optional but recommended: if you enter a secret here, you must pass this value in the `x-pd-secret` HTTP header when making requests
N/A	`http`	`$.interface.http`	This component uses `$.interface.http` to generate a unique URL when the component is first instantiated. Each request to the URL will trigger the `run()` method of the component.
N/A	`db`	`$.service.db`	This component uses `$.service.db` to maintain state between executions.

Trigger Authentication#

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

About Pipedream#

Integration platform for developers

Action#

Choose Best Completion on Cohere

Description:This action chooses the best completion conditioned on a given examples. [See the docs here](https://docs.cohere.ai/reference/generate)

Version:0.0.2

Key:cohere_platform-choose-best-completion

View on GitHub

Cohere Overview#

The Cohere API enables the development of apps with advanced natural language understanding capabilities. Utilizing machine learning, it can help with tasks like text generation, summarization, sentiment analysis, and more. On Pipedream, you can seamlessly integrate Cohere's features into serverless workflows, triggering actions based on text input, processing large volumes of data, or even enhancing chatbots with more human-like responses.

Action Code#

import common from "../common/base.mjs";

export default {
  ...common,
  key: "cohere_platform-choose-best-completion",
  name: "Choose Best Completion",
  version: "0.0.2",
  description: "This action chooses the best completion conditioned on a given examples. [See the docs here](https://docs.cohere.ai/reference/generate)",
  type: "action",
  props: {
    ...common.props,
    prompt: {
      propDefinition: [
        common.props.coherePlatform,
        "prompt",
      ],
      type: "string[]",
    },
  },
  methods: {
    ...common.methods,
    prepareAdditionalData({ prompt }) {
      return {
        prompt: prompt.toString(),
      };
    },
  },
};

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
Cohere	`coherePlatform`	`app`	This component uses the Cohere app.
Prompt	`prompt`	`string[]`	Represents the prompt or text to be completed. Trailing whitespaces will be trimmed. If your use case requires trailing whitespaces, please contact ivan@cohere.ai.
Model	`model`	`string`	Select a value from the drop down menu:`mediumxlarge`
Number of generations	`numGenerations`	`integer`	Defaults to 1, min value of 1, max value of 5. Denotes the maximum number of generations that will be returned.
Max Tokens	`maxTokens`	`integer`	Denotes the number of tokens to predict per generation, defaults to 20. See BPE Tokens for more details. Can only be set to `0` if `return_likelihoods` is set to `ALL` to get the likelihood of the prompt.
Preset	`preset`	`string`	The ID of a custom playground preset. You can create presets in the playground. If you use a preset, all other parameters become optional, and any included parameters will override the preset's parameters.
Temperature	`temperature`	`string`	Defaults to `0.75`, min value of `0.0`, max value of `5.0`. A non-negative float that tunes the degree of randomness in generation. Lower temperatures mean less random generations. See Temperature for more details.
K	`k`	`integer`	Defaults to `0`(disabled), which is the minimum. Maximum value is `500`. Ensures only the top `k` most likely tokens are considered for generation at each step.
P	`p`	`string`	Defaults to `0.75`. Set to `1.0` or `0` to disable. If set to a probability `0.0 < p < 1.0`, it ensures that only the most likely tokens, with total probability mass of `p`, are considered for generation at each step. If both `k` and `p` are enabled, `p` acts after `k`.
Frequency Penalty	`frequencyPenalty`	`string`	Defaults to `0.0`, min value of `0.0`, max value of `1.0`. Can be used to reduce repetitiveness of generated tokens. Similar to `frequency_penalty`, except that this penalty is applied equally to all tokens that have already appeared, regardless of their exact frequencies.
End Sequences	`endSequences`	`string[]`	The generated text will be cut at the beginning of the earliest occurence of an end sequence. The sequence will be excluded from the text.
Stop Sequences	`stopSequences`	`string[]`	The generated text will be cut at the end of the earliest occurence of a stop sequence. The sequence will be included the text.
Return Likelihoods	`returnLikelihoods`	`string`	Select a value from the drop down menu:`GENERATIONALLNONE`
Logit Bias	`logitBias`	`object`	Used to prevent the model from generating unwanted tokens or to incentivize it to include desired tokens. The format is `{token_id: bias}` where bias is a float between `-10` and `10`. Tokens can be obtained from text using Tokenize. For example, if the value `{'11': -10}` is provided, the model will be very unlikely to include the token 11 (`" "`, the newline character) anywhere in the generated text. In contrast `{'11': 10}` will result in generations that nearly only contain that token. Values between -10 and 10 will proportionally affect the likelihood of the token appearing in the generated text. Note: logit bias may not be supported for all custom models.
Truncate	`truncate`	`string`	Select a value from the drop down menu:`NONESTARTEND`

Action Authentication#

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

Generate and copy your API key from your Cohere Platform dashboard.

About Cohere#

Making NLP part of every developer's toolkit. Harness the power of language understanding. Join the developers and businesses who are using Cohere to generate, categorize and organize text at a scale that was previously unimaginable.

More Ways to Connect Cohere + Pipedream#

Other Popular Integrations#

Generate Text with Cohere API on New Scheduled Tasks from Pipedream API

Pipedream + Cohere

Try it

Classify Text with Cohere API on New Scheduled Tasks from Pipedream API

Pipedream + Cohere

Try it

Summarize Text with Cohere API on New Scheduled Tasks from Pipedream API

Pipedream + Cohere

Try it

Chat with Cohere API on New Scheduled Tasks from Pipedream API

Pipedream + Cohere

Try it

Popular Triggers#

New Scheduled Tasks from the Pipedream API

Exposes an HTTP API for scheduling messages to be emitted at a future time

Try it

New Upcoming Event Alert from the Pipedream API

Emit new event based on a time interval before an upcoming event in the calendar. This source uses Pipedream's Task Scheduler. See the documentation for more information and instructions for connecting your Pipedream account.

Try it

Card Due Date Reminder from the Pipedream API

Emit new event at a specified time before a card is due.

Try it

New Upcoming Calendar Event from the Pipedream API

Emit new event when a Calendar event is upcoming, this source is using reminderMinutesBeforeStart property of the event to determine the time it should emit.

Try it