Dialogflow Inbox operates Dialogflow Agents inside a mail server
- Automatic language detection
- Session, thread detection
- Reply recognition
- No HTML parsing
- No component rendering (only text is sent)
- Connect your Agents to a Dialogflow Gateway implementation (more here)
- Install SMTP and IMAP server on your machine, i recommend docker-mailserver
- Point DNS MX record to the SMTP and IMAP server
- Create a user, which will recieve E-Mails for your agent
- Assign alias: your-google-cloud-project-id@yourdomain or *@yourdomain (if you have multiple Agents) to the user
See k8s for examples
- Confirm your Mailserver operates correctly
- Clone the repository
- Install Python and cron
- Install the requirements with
pip install -r requirements.txt
- Configure the environment variables (below)
- Send a test E-Mail to your-google-cloud-project-id@yourdomain and run the inbox.py script with
python
- Setup a cronjob to run it automatically over a interval
Environment Variable | Description | Value |
---|---|---|
INBOX_USER | E-mail user | - |
INBOX_PASSWORD | E-mail user's password | - |
INBOX_HOST | SMTP and IMAP server hostname | - |
FALLBACK_LANG | Fallback language if language detection fails | en |
ENDPOINT | Dialogflow Gateway Endpoint. * for wildcard |
https://*.core.ushaflow.io |
CATCHALL | Catch-all address | - |