OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Chat & Assistants

DoppelBot

DoppelBot is a Slack app that fine-tunes a Llama 3.1 language model on a selected user's Slack messages and uses it to generate replies in that user's style. The repository says message collection, fine-tuning and inference run serverlessly on Modal.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

DoppelBot — OpenViST project profile in Chat & Assistants.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use DoppelBot when a Slack workspace wants replies generated in the style of one selected user's messages, or when a developer wants a documented Modal-based starting point for that workflow.

Who it is for

  • Slack workspaces that want a bot to draft replies modeled on one selected member's communication style.
  • Developers who want to deploy or adapt the Python Slack application on Modal.

Documented uses

  • Choose a Slack user as the style target and start model training with `/doppel `.
  • Check whether the training process has finished by repeating the command.
  • Summon the trained bot by mentioning it in a public Slack channel.
  • Deploy the application using a Slack app, Modal credentials and the repository's documented configuration.
Before you choose

Requirements & limits

  • A workspace can select only one target user, and the selection cannot be changed after installation.
  • Training duration is described as a few minutes to an hour, depending on how prolific the selected user is on Slack.
  • For the installed-app workflow, the README says Modal retains training states but not the Slack messages used for training.
  • The repository is public and carries an MIT license.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Install and use DoppelBot in a Slack workspace

Select one workspace user for training, check training status, then summon the bot in a public channel.

Before you start

  • A Slack workspace where the app can be installed.

Steps

  1. Install DoppelBot in the Slack workspace using the app installation link.

  2. In a channel, run /doppel using the target person's Slack handle or real name.

  3. Wait for training to finish; run the command again to check its status.

  4. Mention the bot in a public channel to summon it.

Expected outcome

A Slack bot that generates replies in the selected user's style.

Guide limitations

  • Each workspace is limited to one target user, and the choice cannot be changed after installation.
  • Training time varies from a few minutes to an hour depending on how prolific the target user is on Slack.
Documentation 1 ↗
Source-based practical guide

Deploy a single-workspace instance on Modal

Set up and deploy the repository's Slack app for one workspace.

Before you start

  • A Modal account, Python environment with the Modal client installed, and Modal token configured.
  • A Slack app in the target workspace with the documented bot-token scopes.
  • A Modal Slack secret containing SLACK_SIGNING_SECRET and SLACK_BOT_TOKEN.

Steps

  1. Create the Modal account, install the modal client in the Python environment, and configure a token with modal setup.

  2. Create a Slack app and add the documented bot-token scopes: app_mentions:read, channels:history, channels:join, channels:read, chat:write, chat:write.customize, commands, users.profile:read, and users:read; install it to the workspace.

  3. Create a Modal Slack secret named slack-finetune-secret with the Slack signing secret under SLACK_SIGNING_SECRET and bot OAuth token under SLACK_BOT_TOKEN.

  4. From the repository root, run modal deploy src.bot; use the printed URL for Slack Event Subscriptions and the /doppel slash command, then install the app to the workspace.

Expected outcome

A deployed Modal app connected to a Slack app for a single workspace.

Guide limitations

  • This deployment path covers one workspace; multi-workspace distribution requires additional configuration and a Neon database.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
13 October 2026
Original website title
modal-labs/doppel-bot · Official homepage title · retained in its source language
How we review sourcesSuggest a correction