Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/SimplyTechnologies/messenger-bot-wit-boilerplate
Facebook Messenger chatbot boilerplate on node/express/mongodb/redis powered by wit.ai
https://github.com/SimplyTechnologies/messenger-bot-wit-boilerplate
boilerplate boilerplate-template chatbot facebook-messenger-bot messenger-bot mongodb witai
Last synced: 3 months ago
JSON representation
Facebook Messenger chatbot boilerplate on node/express/mongodb/redis powered by wit.ai
- Host: GitHub
- URL: https://github.com/SimplyTechnologies/messenger-bot-wit-boilerplate
- Owner: SimplyTechnologies
- License: mit
- Created: 2017-02-02T12:52:00.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2017-02-07T06:24:06.000Z (over 7 years ago)
- Last Synced: 2024-01-16T20:16:16.674Z (5 months ago)
- Topics: boilerplate, boilerplate-template, chatbot, facebook-messenger-bot, messenger-bot, mongodb, witai
- Language: JavaScript
- Size: 14.6 KB
- Stars: 23
- Watchers: 6
- Forks: 10
- Open Issues: 0
-
Metadata Files:
- Readme: readme.md
- License: LICENSE
Lists
- awesome-ai - Facebook Messenger wit.ai node.js boilerplate - Facebook Messenger wit.ai node.js boilerplate (Development / Bot Development)
README
# Facebook Messenger Chatbot Boilerplate
Boilerplate project for developing Facebook Messenger Bot powered by Wit.ai, MongoDB and Redis as a session store.
This boilerplate will allow you to bootstrap your chatbot within 8 minutes, all you need is wit.ai application setup, Facebook page, Facebook app and running MongoDB and Redis.
## External dependencies
### Wit.ai
The AI engine used in this boilerplate is wit.ai, please make sure to create a new app in the [wit.ai](https://wit.ai/) dashboard.
### MongoDB
This boilerplate is using MongoDB for storing user's data and for tracking their last activity with the bot
Please install MongoDB v3 or greater and setup connection in the `config/config.js`
By default the in the development environment `chatbotdb` database will be created
### Session store using Redis
Redis server is required for managing sessions and keeping the context with each user.
Please install Redis and setup connection in the `config/config.js`
## Setup
`npm install`
## Configuration
Set environment variables or write values directly in the `config/config.js`
`FB_PAGE_TOKEN` - Access token generated for the Facebook application to allow making Graph API requests to the page
`WIT_TOKEN` - WIT server side token
`FB_WEBHOOK_VERIFY_TOKEN` - a verify token that should be used for Webhooks setup, this is something you need to generate by yourself
`FB_PAGE_ID` - ID of the Facebook page that will be used as a bot.
## Run
In order to run and test chatbot it is required to have server running with secured connection. For that you can use **ngrok**
###### install ngrok and run
`./ngrok http 3000`
###### run the server
`npm start`
to enable debug mode
`DEBUG=* npm start` or `DEBUG=cbp* npm start` for app level log messages only
###### update Facebook application's webhook
`https://xxxxxxxx.ngrok.io/bot`
enter the `FB_VERIFY_TOKEN` value and submit
## Project structure
```
actions/
- define custom wit.ai actions here, the file name should be the name of the action, all actions from this folder are automatically registered as wit.ai actions.
config/
- configuration files for access keys, tokens, connection strings
handlers/
- messenger event handlers for text messages, attachments, postbacks and quick replies
init/
- init scripts for mongodb and redis
schemas/
- mongodb/mongoose schemas
scripts/
- shell scripts for messenger greeting text, persistant menu and getting started button setup
services/
- custom services
app.js
- application entry point
graphAPI.js
- Facebook Graph API helper
platformHelpers.js
- Helpers for generating message templates, e.g send location, quick replies and others
routes.js
- express routes setup and message handler delegation
sessionStore.js
- Redis based session store
wit.js
- wit.ai setup with actions loading
witHelpers.js
- helpers for extracting entities from wit.ai response
```
## The Roadmap
- [x] Redis session store
- [x] Automatic actions binding
- [x] User info tracking in the database
- [x] Saving last and previous geo locations shared by user
- [x] Demo of send location, quick replies
- [x] Shell scripts for adding greeting message, persistent menu and getting started button
- [ ] Demo of generic template
- [ ] Demo of list template
- [ ] Unit tests
---
[ChatBotKitchen](https://www.chatbotkitchen.com) project
*made with love ❤️ in Armenia by [Simply Technologies](http://www.simplytechnologies.net)*