You can now customize the AI prompts used to generate Reddit posts directly from the Post Generation page.

What's New:

- Collapsible Prompt Editor - Access prompt settings via expandable section on Post Generation page

- Dual Provider Support - Separate prompts for Anthropic (Claude) and OpenAI (GPT)

- Full Customization - Edit the entire prompt template including guidelines, tone, and instructions

- Variable System - Required variables (brand info, subreddit, rules, examples) are validated before saving

- Reset to Default - One-click reset to restore original prompt at any time

- Owner-Only Access - Only project owners can modify prompts; team members can view

How to Use:

1. Go to Post Generation page

2. Click "Post Generation Prompt" to expand

3. Select provider tab (Anthropic/OpenAI)

4. Edit the prompt template

5. Click "Save Changes"

Receive real-time notifications whenever opportunities are discovered or responses are generated. Integrate ReplyDaddy with your existing tools and workflows.

- Webhook Configuration - Set up a custom webhook URL in Settings to receive POST requests when events occur

- Two Event Types - Get notified on opportunity.found (new leads discovered) and response.generated (replies created)

- HMAC Security - Optional secret key for payload signing with SHA-256 to verify webhook authenticity

- Test & Preview - Send test webhooks and view sample payloads before going live

- Full Data Payload - Each webhook includes complete opportunity data, scores, matched keywords, generated response, and persona information

- No Duplicates - Built-in tracking ensures each event is only sent once per opportunity

Use Cases:

- Push new leads to your CRM or Slack

- Trigger automated workflows in Zapier/Make

- Build custom dashboards with real-time data

- Archive opportunities to your own database

Available to project owners in Settings > Webhook Notifications

On the oppurtunity card now you can choose to mark an oppurtunity as viewed, or mark as replied.

You can keep them viewed if you want to reply later, or they are irrelevant.

With Role Based access being live

• Now each user can have their unique persona -> Invite n number of users to manage -> n number of reddit accounts -> All reply & post generation will be unique based on the persona!

Here are the allowed permissons:

• Invitations: Owners invite members per project via email link (Google OAuth signup)
• Personas: Each member creates their own persona per project; scanning and generation uses their persona
• Shared limits: Keywords, scans, and posts are pooled at the project level
• Individual generation: Reply and post generation is unique per user based on their persona

In some cases when the user had multiple projects, due to cached data, the project data for different projects was visible in under wrong projects, causing confusion and giving a look of data corruption, however all project data is perfectly stored against correct projects, only due to caching on the frontend these discrepancies were there which have been resolved!

If you have a new reddit account, less than 30 days old and below 50 Karma, a warning is added on post generation, even though your account might be eligble as per rules, there are hidden rules, which will get your post removed most of the times

The old Settings & Usage Page have been divided into two disctinct pages.

1. Persona Page -> Where the users can manage the persona for the project
2. Settings Page -> Where users can manage the project settings, such as discovery settings, API keys, and track the usage stats

We just pushed a UX update, but it resulted in a critical bug on production where project data is not being loaded correctly, the issue has been found and a fix is being implemented, ETA 15-20 Mins.