Roadmap
Changelog

Share your feedback, suggest features, and report bugs!

Tell us what we can do to make Chatling the best product for you.

Organized Knowledge Folders & Deep-Crawl Automation

Title: Web-Sync Folders and Deep Crawling for Embedded Third-Party Data (IDX, MLS, eCommerce, Booking Engines) The Problem: Management Fatigue: Managing hundreds of synced links in a single list is unscalable. Users need to categorize links into folders to update or audit specific sections of their site without manually searching the entire database. Hidden Data: Many businesses use "Search & Filter" services that embed content via JavaScript or iframes. Standard sitemap crawlers often miss these links, leaving the AI "blind" to the most critical business data (like inventory, listings, or menus). Target Industries & Use Cases: Real Estate (IDX/MLS): Crawling property listing links embedded via IDX Broker or similar services. Automotive (Inventory Feeds): Scraping car detail pages (VDPs) that are pulled into a site from a third-party DMS (Dealer Management System). Travel & Hospitality: Syncing dynamic booking pages for hotels, tours, or vacation rentals embedded from central reservation systems. eCommerce & Directories: Capturing product pages or affiliate listings that are dynamically generated and not listed in the primary static sitemap. The Proposed Solution: Categorized Web-Sync Folders: Allow users to create folders (e.g., "New Inventory," "Archived Services," "Product Catalog") to organize URLs. This allows for bulk actions, like "Sync only the 'Current Sales' folder.""Deep-Link" Crawling: An advanced crawler option that scans a page’s code to find and follow links generated by embedded scripts or third-party widgets, even if they are missing from the sitemap.xml. Smart Auto-Sync: Automatically detect when these embedded links change or expire, keeping the AI knowledge base current without manual intervention. Business Impact: This feature would transform Chatling from a static website bot into a powerful, real-time data agent for any industry that relies on dynamic inventory. It reduces "data rot" and ensures the AI provides accurate, up-to-the-minute information to customers.

larusso About 14 hours ago

💡 Feature Request

Learning/Training-Actions

Objective: To introduce a dedicated Description & Procedure field within individual AI Action blocks to allow for granular guidance on how a task should be executed. The Current Gap: Currently, training the AI on specific procedures is often buried in general "AI Instructions." This can lead to the AI missing nuances when it triggers a specific action or task. The Proposed Solution: Add a rich-text field within the Action configuration (e.g., Lead Forms, HTTP Requests, or Custom Buttons) called "Task Guidance." This field would allow users to: Define step-by-step logic specific only to that action. Provide contextual details (e.g., "Ask for X before performing Y"). List constraints for the procedure (e.g., "Only proceed if the user has provided a valid order number"). Why this is effective: Precision: It ensures the chatbot follows specific company protocols without cluttering the global instructions. Scalability: It makes managing complex workflows easier as the logic is attached directly to the task it governs. Reliability: It reduces "hallucinations" by giving the AI a clear manual for every triggered action. Example Use Case: Action: Refund Request. Procedure Detail: 1. Check the knowledge base for the 30-day policy. 2. Verify the user has their receipt. 3. If both are met, provide the link to the form. Otherwise, offer a 10% discount code found in doc_v2.

larusso About 14 hours ago

💡 Feature Request

We will Upgrade to Enter Enterprise ($99) plan if you fix this bug!

Right now, when I add a button (for example, “Back to Main”), the text input option disappears from that block. That means if a user wants to type instead of clicking the button, I have to create and connect a separate “User Input” block. From a UX point of view, this feels limiting. Buttons should just be quick suggestions — users should still be able to type freely in the same step. Most modern chat experiences allow both. The current setup makes flows more complex than they need to be and doesn’t feel very natural in conversation. It would be great if buttons and free text input could work together in the same block — button clicks routed normally, and typed responses handled as fallback or intent-based input. This would really improve how we build flows. If this is something you can enable for us, we’re happy to upgrade from the Basic plan to the Enterprise ($99) plan.

Barath Sivanesan 22 days ago

💡 Feature Request

Greeting Message section - agent icon shown only on last message when multiple messages are added

Description: In the Greeting Message section ("The messages that will be shown at the beginning of the chat"), when adding multiple greeting messages, the chat agent icon appears only next to the last message. The earlier messages do not display the agent icon. This creates a visual inconsistency in the chat UI, as the first messages appear misaligned compared to the last one. Steps to reproduce: Go to the Greeting Message section Add more than one greeting message Open the chat preview Actual result: Only the last greeting message shows the chat agent icon Expected result: Either: the agent icon should be displayed next to all greeting messages, or the layout should include consistent spacing/alignment where the icon would normally appear (similar to messaging platforms like Facebook Messenger) Additional notes: The current behavior feels visually inconsistent and may confuse users Screenshots attached for reference AI Agent: Facbook Messenger example layout:

J.A.R.Wiz 25 days ago

1

🐛 Bug Reports

Richiesta indirizzo email

vorrei evitare di chiedere l’email dell’utente per accedere alla chat perchè a mio avviso questa richiesta non favorisce il dialogo, ma vorrei mettere qualcosa nel builder che ad un certo punto chiede l’email per, ad esempio, proporre all’utente l’invio della conversazione (che ovviamente dovrei fare io in modo manuale): è possibile?

Luigi Nardullo About 1 month ago

💡 Feature Request

More customisation of design with corner rounding + chat thinking

Could we get the ability to 1. better customise the font of the response 2. perhaps create an account page with memory of user’s conversations 3. customise the … to custom text like Thinking… / our own text whilst the AI is loading 4. customise corner rounding to make it more rounded 5. customise background colour

Pack About 2 months ago

4

💡 Feature Request

Completed

chatling ai credit limit

chatling lets be honest, running businesses on 3000 credits or even 12000 credits isnt enough lemme explain. if you run 5 chatbots for various businesses lets say medium traffic it might not be enough even when we are talking abt combined i think we should raise the ai credit limit a bit more for easy business handling and easy customer support thank you chatling.

Faizan Ahmed 2 months ago

1

💡 Feature Request

Completed

Blocking Issue: Triggering Flow via API & Accessibility Roadmap Update

Hi Chatling Team, We are developing a custom chat frontend using your API v2 to adhere to strict WCAG accessibility standards, as the native widget is not yet compliant. We are facing two major issues: 1. Technical Blocker (API v2): We are unable to trigger a Flow (Playbook) via the API. Even when sending specific trigger keywords associated with an Intent, the API ignores the Flow logic and returns a generic Knowledge Base response. Question: Is there a specific parameter or method to force the API to bypass the KB and execute a structured Flow? 2. Accessibility Roadmap: We raised the issue regarding the native widget's accessibility a few months ago, and it was added to your roadmap. Question: What is the current ETA for full accessibility support in the native widget? Looking forward to your quick guidance. Best regards, Koby

koby 2 months ago

5

🐛 Bug Reports

Completed

An error occurred, please retry. (CODE: 3298)

Hello Chatling Support, I have a paid plan on Chatling and successfully connected my Instagram business account. However, every time I try to send or receive messages, I get the following error repeatedly: An error occurred, please retry. (CODE: 3298) I have: Verified that the Instagram account is a Business Account with correct permissions. Confirmed that the token is valid and not expired. Checked the Webhook URL is HTTPS, publicly accessible, and with valid SSL. Despite this, the error persists. It seems to be a server-side issue, not related to my setup. Please check the server logs for this integration, as this is critical for my workflow. I need this fixed urgently since I am on a paid plan and cannot use Instagram integration. Thank you for your prompt assistance. Best regards,

monscorpdevz 2 months ago

5

💡 Feature Request

Smaller / simpler appearance on mobile

It would be nice to have a few options for the bot icon on mobile devices: the ability for the user to completely hide the icon or minimize smaller (something on the edge). It is fairly distracting with the default icon on a mobile device given the very limited space. the option to never show on mobile (I can modify the script to not load on mobile, but would be handy to have a setting in Chatling or for the default script tag to have a boolean falg to do this) some ability to tweak settings differently for mobile - in particular I’m setting very small offsets to get out of the way on mobile, but the default spacing looks better on a computer.

greg98236423 3 months ago

💡 Feature Request

Agents need more control

I’d like to use an agent on Instagram, but ONLY when it can provide useful information. Otherwise, the message should not get a reply (I’ll do it manually). There appear to be no way to give an agent instructions so you can give it guidance, which would be very helpful to make sure it only responds to situations where you would want it to. And there is no option to add a way to clarify that a bot was responding (such as some simple text I could prepend to the first reply, such as “AI bot:” to help them understand they are interacting with a chatbot and not me).

greg98236423 3 months ago

💡 Feature Request