We're excited to announce Telegram as a new support channel in Gleap! You can now connect a Telegram bot and manage all customer conversations directly from your Gleap inbox - just like WhatsApp, Instagram, and Facebook Messenger.

What's included:

• Receive and reply to Telegram messages from your Gleap inbox

• Full media support - images, documents, videos, voice notes, and stickers

• AI bot and workflow support with interactive inline buttons

• Typing indicators shown to customers while agents compose replies

Getting started:
Create a Telegram bot via @BotFather, paste the token in Settings → Channels → Telegram, and you're live in under 2 minutes.

You can now power Kai with GPT-5.4 - OpenAI's latest and most capable model yet.

Expect sharper, more accurate responses with significantly fewer hallucinations, improved agent skill execution, and faster response times across the board. A meaningful upgrade for any team looking to get the most out of their AI support experience.

Head to your Kai settings to switch models and give it a try (AI agent → Settings → Models).

Reaching people outside your existing user base just got a whole lot simpler. You can now upload a CSV file and send outbound emails directly to your imported contacts — no manual entry, no workarounds.

How it works:

• Head to Outreach → New Email and choose your campaign content.

• Under Targeting → Audience, select CSV Upload.

• Map your CSV fields (like email, name, or any custom data) to the right session properties.

• Click Upload to create sessions for each contact, then hit Send Now to deliver your email.

This was one of your most requested features — and we're glad it's finally here. Whether you're running an outreach campaign, re-engaging a list from another tool, or reaching out to event attendees, CSV upload makes it fast and flexible.

Head to your Gleap dashboard and give it a try.

You can now connect your Notion workspace directly as a content source for your Gleap AI bot. Import your existing pages, wikis, and documentation in just a few clicks — no manual copy-pasting required.

Once connected, you have full control over how your content is used. Choose to make all imported pages publicly available to the AI, or respect your existing Notion permissions so that private pages remain internal-only. You can also decide whether your content should stay automatically in sync with your Notion workspace, or if you prefer to trigger updates manually whenever you're ready.

It's the easiest way to turn your existing Notion documentation into a powerful knowledge source for your AI bot.

Read more here

Kai can now index your codebase and use it to directly support your customers. Connect a repository and Kai will understand your application, so it can guide users through your product without needing manually written help articles for every detail.

This is especially powerful for frontend applications. Kai can help customers find settings, walk them through features, and explain how things work — all based on your actual code.

What's new:

• Connect a Git repository to let Kai index your code

• Kai uses your codebase as a knowledge source when answering customer questions

• Customers get accurate, up-to-date guidance that reflects your actual product

Use cases:

• "How do I change my notification settings?"

• "Where can I update my billing info?"

• "How do I enable dark mode?"

• Kai locates the answer in your code and guides the customer step by step

Why this matters:
Your product evolves faster than your documentation. Code indexing lets Kai stay in sync with your latest shipped features automatically, reducing the need to constantly update help docs and knowledge bases.

Good to know:
Indexing consumes AI tokens and can take up to an hour depending on the size of your codebase. To give you a sense of scale — indexing our own project cost around $30 in AI tokens and produced roughly 15,000 knowledge points.

Get started:
Head to AI Agent → Content → Code Repositories to connect your first repository.

Keeping your feature request board clean just got a whole lot easier. We've added a new AI-powered tool that automatically finds and merges duplicate feature requests — so you can spend less time tidying up and more time building what your users want.

How it works:

Head to Feedback Portal → AI Tools → Deduplicate items and hit Scan for duplicates. Our AI will analyze all your open feature requests — comparing titles, descriptions, and context — to surface groups of duplicates you might have missed.

For each group, you'll see the matching requests side by side with their upvote counts and statuses. When you're ready, hit Merge and we'll take care of the rest:

• A polished, customer-facing title and description are generated automatically

• Upvotes from all duplicates are consolidated into one

• Comments are preserved and moved to the merged request

• Duplicate requests are archived to keep your board clean

Results are cached so you can come back anytime without re-running the scan. When you're ready for a fresh analysis, just hit Rescan.

We're excited to launch the Feedback Portal, a single, beautifully designed space where your users can actively shape your product.

From one place, users can now submit and upvote feature requests, comment on ideas, browse your public roadmap, and follow along with every release through an interactive changelog — complete with emoji reactions, comments, and email subscriptions.

For your team, it means less "is this coming?" noise in support, clearer prioritization based on real demand, and a transparent way to close the feedback loop when something ships.

Head to your Gleap dashboard to set it up and make it your own.

Read more here