Changelog

Meet Pluno for your browser

Pluno is now available as a browser extension that gets things done in any web app, 10x faster than screen-clicking agents. Instead of navigating UIs one element at a time, Pluno talks to each site's API directly, so it stays fast and doesn't break when a button moves. It comes pretrained on 50,000+ tools and learns any new one you show it in seconds.

Tell it the outcome, like "turn off email digests across Notion and Slack", "pull last month's Stripe payouts into a CSV", or "reassign 200 HubSpot contacts to a new owner", and it handles the rest. You approve every action before it runs, and it improves with every interaction through self-learning.

To try it out, add Pluno from the Chrome Web Store. It works on Chrome, Edge, and other Chromium browsers. New signups get free credits, and you can earn more by referring friends or sharing Pluno on social media. Learn more at pluno.ai/browser.

The Troubleshooting Agent is now available

The Troubleshooting Agent can now be tried self-serve: sign up, get free trial credits automatically, and upgrade to the usage-based plan whenever you're ready. Checkout, credit balance, and top-ups are all handled in-app.

Highlights since launch:

• Works across Slack, the web app, Zendesk, and Intercom. In Slack it keeps context in threads, reads quoted and linked messages, can search across your workspace's channels, and stays silent when a conversation doesn't need it.
• It can now create GitHub pull requests: when it's confident in both the diagnosis and a code fix, it offers to open a PR, and only does so after your explicit approval. Agent branches are prefixed with pluno/ so they're easy to spot.
• It learns from your sessions: skills are automatically created and refined based on real usage, so troubleshooting gets more accurate over time.
• A new Usage dashboard shows credits consumed, sessions, and tokens over time, broken down by source, with full session transcripts.
• Set up connectors in a single unified flow with suggestions for the most common tools. All integrations connect read-only, and connected Zendesk, Jira, and Knowledge Base integrations are picked up automatically. You can invite team members so different people can manage different connectors.
• Guide its behavior with custom instructions in the agent settings, share images and files in chat, and enjoy shorter, cleaner replies.

Linear is now supported for escalations

The Escalation Copilot now works with Linear, with full feature parity with Jira. You can create, link, and unlink Linear issues from both the Zendesk and Intercom sidebars, send comments (cmd/ctrl + enter works too), and Pluno listens to Linear comments automatically. Multiple Linear workspaces are supported.

Also included:

• Escalation templates with Linear-specific fields, plus team and project restrictions per Zendesk brand.
• A Pluno view inside Linear: configure which Zendesk fields are shown on linked issues, with aggregation for numeric fields.

Get started at app.pluno.ai/escalation-copilot/linear.

Pluno Copilot now works with Intercom

The AI Copilot and Escalation Copilot are now available for Intercom via the Pluno Copilot Chrome extension. Connect Intercom from the dashboard and you get the same suggested answers, escalations, and knowledge features you know from Zendesk.

• Suggested answers are shown in your support team's language for non-English conversations.
• Choose which Intercom admin account authors the internal notes Pluno posts (e.g. a dedicated "Pluno" admin).
• Track Pluno usage directly in Intercom attributes to measure its impact.

Other fixes and improvements

• Deflection AI now uses the previous ticket's context when a ticket is reopened, and automatically merges duplicate tickets from the same requester.
• Pluno AI now handles follow-up tickets to previously escalated tickets.
• When Pluno merges Zendesk tickets, the customer is now also notified on the source ticket.
• Automated answers are no longer interrupted by internal notes.
• Fixed an issue where the AI agent escalated too often and could get stuck repeating responses.
• Fixed deflected tickets that could get stuck in pending without being handed off or auto-resolved, and deflection messages that could fail silently after too many retries.
• Long AI replies on Zendesk messaging conversations (web widget, WhatsApp, social) are no longer dropped; they're now shortened and delivered instead of failing.
• Non-public Help Center articles are now used in Copilot only by default, with an explicit opt-in to also use them in the deflection agent. Public articles are unaffected.
• Improved chunking of Help Center articles for more accurate knowledge retrieval.
• You can now see the Copilot's debug output (suggested answer, agent guidance, and cited references) in the ticket chat history, so you can improve your internal articles over time.
• Copilot usage on Zendesk can now be tracked via per-ticket tags (escalations, copilot chats, suggested-message inserts). Enable it under Settings → Sidebar Preferences.
• The ticket summary in the sidebar now keeps showing the previous version while it regenerates, instead of disappearing on every new message.
• When chatting with a Copilot session generated before new messages arrived, Pluno now adds context about the time passed and the new messages.
• Zendesk tickets can now be attached to existing Slack threads, not just ones Pluno created.
• Escalation templates: the priority field is now at the top of Jira escalation forms, and subsequent escalations after the first now generate an acknowledgment message.
• Fixed empty escalation content when a field had no allowed values, and fixed the linked Jira issues card blinking in the sidebar (with faster loading during escalation).
• The "Insert into editor" button is now more prominent, and the "Send feedback" link was removed from the ticket editor and top bar to avoid confusion with ticket replies (it remains in the sidebar).
• The ticket popup in the Zendesk sidebar can now be permanently dismissed, and the "Process ticket" button works reliably again.
• QA scoring weights and point deductions are now configurable per criterion.
• Detailed QA results can now be exported to Excel, including per-criterion results and final scores per ticket, and exports now show consistent, real Zendesk agent names instead of emails or aliases.
• The QA "Excellent" badge now requires a score above 95%, and QA evaluator comments are now written in your support team's language.
• Customers are now notified by email when their monthly AI Field Filling limit reaches 50%, 80%, and 100%.
• Shopify can now be connected via the official OAuth flow, which is now the default way to integrate.
• Confluence ingestion now preserves special characters like German umlauts.
• The dashboard now warns you if multiple Pluno apps are installed, with the option to deactivate all but one.
• Users with separate Discord and Zendesk communities now see the correct dashboard for each.

Special thanks to our users MaintainX, Lightspeed, Innovorder, Velotric, HelloBetter, Bluetti, and Daparto for their feedback that inspired these updates.

Pluno now integrates with Shopify

Support teams using Shopify often lack real-time order context when helping customers, leading to slower resolutions and repetitive back-and-forth.

When connected, Pluno automatically pulls relevant customer data and history from your integrated systems, giving your AI the full context it needs to assist effectively across Deflection, Copilots, and Tagging.

To get started, contact us and our team will set up your Shopify integration for you.

Ticket form selection for AI Tagging & Field Filling

You can now include ticket form selection as part of the AI Tagging & Field Filling workflow. From the Pluno dashboard, you can configure ticket forms just like any other field. When the workflow runs, Pluno selects the appropriate ticket form first, then fills in the fields based on that form, so the AI works with the right structure from the start and applies values more accurately across different support flows.

Other fixes and improvements

• You'll now receive an email when Pluno training finishes after your imported historical tickets have been processed.
• The agent onboarding modal in Zendesk is now shown only after training has finished.
• Fixed an issue where "allow learning from this & newer tickets" did not include the current ticket.
• Fixed an issue that prevented learning from past tickets from succeeding for some customers.
• Agents are now prompted with Pluno onboarding inside Zendesk after the app is installed on their subdomain.
• Fixed an issue where the sidebar still showed the onboarding/setup button after onboarding was completed.
• The sidebar now shows a clear error when the Pluno API key is missing or invalid.
• Zendesk triggers have been consolidated into a single trigger, reducing both trigger count and Zendesk requests.
• Fixed an issue where Escalation Copilot sometimes wrote internal team updates in the wrong language.
• Fixed an issue where Escalation Copilot sometimes mixed up context across multiple linked Jira issues.
• Fixed an issue where Copilot could not be retriggered if it was stuck for too long.
• Fixed an issue that caused missing escalations and duplicate email collection in the chat widget.
• Pluno now remembers your last selected Jira project, issue type, and Slack channel for escalations across sessions.
• Escalation attachments are now filtered to exclude email signature images.
• Fixed an issue where the AI Field Filling plan badge was shown for non-paid plans.
• Clarified AI Tagging instructions and labels to reduce confusion around comma-separated tags.
• Fixed a race condition where Deflection could overwrite tags created by Tagging runs.
• You can now disable email collection for the AI agent in Zendesk Messaging.
• You can now reorder collected fields using drag and drop.
• Updated the email collection message to make it clearer that providing an email is required to connect with the team.
• Fixed an issue where user messages were missing from tickets when they came from a different email address.
• You can now add additional instructions and context for ticket summary generation under Settings > Sidebar Preferences.
• Fixed an issue where Pluno did not account for unreadable attachments; prompts now mention that an attachment exists even when its contents cannot be accessed.

Special thanks to our users Velotric and MaintainX for their feedback that inspired these updates.

Custom API Integration is Now Available

You can now connect external APIs as tools for your Copilot and deflection agents. This gives your AI the ability to fetch live data or trigger actions in external systems mid-conversation, so instead of escalating or asking customers to check elsewhere, it can look up the answer directly. The result is fewer handoffs, faster resolutions, and a support experience that feels genuinely intelligent.

To try it out:

1. Reach out to us with your API documentation and we'll get it configured for you.
2. Once set up, your AI agent will automatically draw on your system data when responding.

Notion is Now Available as a Knowledge Source

You can now connect your Notion workspace as a knowledge source for Copilot. The integration lives under Knowledge Hubs and is available on a per-brand basis, so each brand in your account can have its own Notion connection.

How to try it out:

1. Go to Knowledge → Knowledge Hubs and select the brand you want to connect.
2. Click Notion in the left-hand menu.
3. Click + Connect Notion Workspace and authenticate with your Notion account.
4. Once connected, choose which Notion pages to use as a knowledge source, select specific pages or the entire workspace, whatever fits your needs.

Other fixes and improvements

• Added an info tooltip in the Pluno reporting dashboard explaining how AI resolved is counted, making it easier to understand your resolution metrics at a glance.

Special thanks to our users Waste Vision, Cobb Tuning, Velotric, Venga, Happeo, Alterra, Innovorder and MaintainX for their feedback that inspired these updates.

Sidebar Content can now be Displayed in your Team's Language, regardless of the Ticket Language

Support teams can now configure a preferred language for sidebar content that's independent of the ticket's language.

With this update, the following are now rendered in your configured team language:

• Ticket summary
• AI Copilot Agent Guidance & References
• AI Copilot suggested messages are generated in both the user's language and the team's language
• Escalation Copilot internal notes to the human agent are generated in the team's language

To try it out:

Go to Settings → Sidebar preferences

1. Under Support Team Language, open the dropdown ("Language of the ticket" is the current default)
2. Select your team's preferred language
3. Open any ticket,the sidebar content will now reflect your chosen language, while the end-user conversation remains unaffected

Escalation Templates can now be Configured per Jira Project and Slack Channel

You can now configure unique escalation templates for individual Jira projects and Slack threads. Previously, only a single global template could be defined, which applied to all escalations regardless of destination. Teams that escalate to multiple projects or channels can now tailor the template content for each one.

How to try it out

1. Go to Escalation Copilot → Jira Escalation (or Slack Escalation)
2. Under the Escalation Template tab, open the template dropdown, you'll see the existing Default template alongside a new + Add template option
3. Click + Add template and select the Jira project (or Slack thread) you want to configure a template for.

Other fixes and improvements

• The most recent project and issue are now preselected in the form when escalating to Jira.
• File attachments during escalations are size-checked in advance, with warnings if any fail.
• You can now search existing Jira issues for escalation without selecting a project type first.
• When a linked Jira issue is moved, a note is now added in Zendesk.
• Jira user mentions in comments posted from Zendesk now appear correctly as names instead of raw links.
• Escalating or unlinking an issue/thread now posts an internal note in Zendesk with direct links to the related escalation.
• Internal notes now sync successfully between Zendesk and Jira/Slack even when attachments are included.
• Users are now notified when auto AI field filling makes no changes, with unchanged fields clearly listed in the notification.
• Jira projects used for AI Copilot knowledge can now be configured in Escalation Copilot settings.
• Added a fallback so when an agent chats with Copilot and it has no prior context, it builds context from the agent’s message and ticket thread.
• Suggested replies in the AI Copilot now include a notice when they may be outdated, prompting regeneration if needed.
• Quality Assurance AI can now verify if required ticket fields were filled based on criteria you define.
• Fixed an issue where AI field autofill didn’t work properly for form-based tickets.
• AI field filling now supports the Assignee field.
• Fixed a bug where fields excluded in the dashboard were occasionally autofilled in Zendesk.
• AI Copilot crashes now show a clear error message instead of failing silently.
• Reduced unnecessary API calls to lower operational overhead.
• Knowledge items can now be renamed or deleted for easier management.
• Fixed an issue preventing past-ticket knowledge from updating for some users.
• The dashboard now shows badges when AI training is in progress.
• Fixed an incorrect redirect when accessing the knowledge base page.
• Fixed a rate-limit issue affecting some users during Zendesk OAuth.
• Applied multiple fixes to prevent rare Zendesk sidebar crashes.
• Fixed onboarding step two not being marked complete after finishing.
• Corrected inaccurate onboarding status for users returning after long inactivity.

Special thanks to our users MaintainX, HelloBetter, Crexendo, and Innovorder for their feedback that inspired these updates.

Link Multiple Jira Issues and Slack Threads to a Single Zendesk Ticket

You can now link a Zendesk ticket to multiple Jira issues and Slack threads with Pluno. Previously, you could only link a ticket to a single issue or thread in either Jira or Slack.

This allows you to collaborate on a ticket with teams across both platforms and prevents the creation of duplicates. By linking a single ticket to already existing threads and issues, you not only keep everything clean and organized, but you also gain clarity into things like the ticket's severity, how many people have been affected, and the full scope of the issue.

Try it out by using the escalation buttons in the Pluno app inside the Zendesk sidebar. Previously, these buttons didn't show for already escalated tickets.

Optimize Escalation Copilot with Debug Insights

The Ticket Chat History page in the Pluno dashboard now displays detailed debug information for every escalation copilot action. You can see the trigger reason, the copilot's reasoning process, suggested replies, internal notes, and messages sent to your team through Slack or Jira.

Previously, understanding the escalation copilot's decisions and team communications required contacting support, making it difficult to troubleshoot issues or improve the copilot’s behavior using AI workflows.

With complete debug logs, you now have the insights needed to optimize it based on actual execution data rather than assumptions.

To access debug information, navigate to the Ticket Chat History and click on any escalated ticket entry in the sidebar to view the full ticket thread and toggle the Escalation Copilot activity to view the insights.

Other fixes and improvements

• You can now include file attachments when escalating tickets to Slack.
• Added brand-based restrictions for Jira projects, issue types, and Slack channels in escalation routing.
• Jira escalations are now 2x faster and more reliable through optimized processing and improved caching.
• Jira escalations now support the 'Team' field.
• Using your custom escalation template, you can now populate Jira labels and automatically create new labels in Jira if they don't already exist.
• When starting a Jira escalation, the project and issue type now default to your last selection, this previously broke if you escalated to Slack in between.
• Jira linking option is now positioned above "Create Issue" for easier access.
• Now you can see the status of Jira issues when searching to link them to a ticket.
• Improved Jira issue search: you can now find issues like "ABC-123" by searching just "123".
• Jira field loading errors are now displayed in the Zendesk sidebar.
• You can now choose which Zendesk user account the escalation copilot uses when posting internal notes.
• Fixed copilot prompt ticket context: assignee is now included, and ticket fields are now available for messaging tickets.
• Fixed issue where AI copilot messages sometimes had missing newlines.
• QA ratings in the Zendesk sidebar can now be expanded to see the explanation behind each rating.
• Fixed QA loading issue for communities with many agents and improved UI with scrollable components.
• Confluence page listings now exclude personal spaces and only auto-fetch when you have 20 or fewer spaces to improve performance.
• Fixed issue where Confluence integration was cutting off content on large pages.
• Fetching Help Center articles with "Include non-public pages" now works correctly for different brands.
• Chat transcripts from end users are now added as internal notes when they're the first message in a ticket, reducing tickets without messages in the dashboard.
• Zendesk agents now appear in search results on the Pluno dashboard’s Members page invite panel.

Special thanks to our users Velotric and Innovorder for their feedback that inspired these updates.

Ensure accurate answers from human agents with AI

Our AI-powered Quality Assurance feature just got even more powerful. You can now use it to ensure the answers provided by human agents were accurate. This is done by having the AI check the answers against your knowledge base, and then have the it rate their accuracy, and provide actionable feedback.

To use this feature, simply set up a QA criterion in the Pluno dashboard, like “Answer correctness.” The AI then checks, flags inaccuracies or missing info, and assigns a 1-5 score with a brief explanation for improvement.

Our AI-powered Quality Assurance feature just got even more powerful. You can now use it to ensure the answers provided by human agents were accurate. This is done by having the AI check the answers against your knowledge base, and then have the it rate their accuracy, and provide actionable feedback.

To use this feature, simply set up a QA criterion in the Pluno dashboard, like “Answer correctness.” The AI then checks, flags inaccuracies or missing info, and assigns a 1-5 score with a brief explanation for improvement.

Prevent end-user frustration with silent escalations

Sometimes, users are already frustrated when they contact support. If a deflection AI first tries to handle their issue and then says it’s escalating to a human agent, it can make them more annoyed, wondering why it wasn’t escalated immediately.

Now, you can create a deflection workflow that silently escalates specific ticket types to human agents without telling the user an escalation is happening. This is especially useful for high-urgency requests (e.g. billing disputes, security concerns, outages, VIP customers), where a fast human handoff is the best experience.

Sometimes, users are already frustrated when they contact support. If a deflection AI first tries to handle their issue and then says it’s escalating to a human agent, it can make them more annoyed, wondering why it wasn’t escalated immediately.

Now, you can create a deflection workflow that silently escalates specific ticket types to human agents without telling the user an escalation is happening. This is especially useful for high-urgency requests (e.g. billing disputes, security concerns, outages, VIP customers), where a fast human handoff is the best experience.

Other fixes and improvements

• Answer Copilot now runs automatically after agent notes are added to the ticket, and also if a ticket is escalated following a deflection attempt, ensuring it always provides relevant suggestions when context changes.
• The deflection workflow guide has been updated to clarify what workflows can see, when they run, what actions they can take, and what’s not supported.
• Workflow text areas didn’t expand with content while editing, making it harder to see long entries; now they automatically grow to fit the content.
• Call summaries sometimes generated incorrect information for empty recordings; now this issue is fixed.
• In the Pluno dashboard, the Support Agent role description wasn’t clear, and users could still see restricted pages in the navigation; now the description clearly explains what the role entails, and the navbar only shows pages a user can access.
• Before, the Members page on Pluno dashboard didn’t guide users; now it shows helpful messages when empty or when a search has no results.
• Added a tooltip to help new users quickly understand how to set limits on a knowledge base web link, such as restricting access to public pages only.
• All Zendesk fields were listed previously in the AI Field-Filling configuration page, which could be confusing; now only active ones are shown, with an updated description for clarity.
• Previously, onboarding didn’t allow switching Zendesk subdomains easily; now a small button lets users connect to a different subdomain during onboarding.
• The second onboarding step (API key) was incorrectly marked as completed if a wrong key was installed; now it correctly requires the user to complete the step.

Special thanks to our users PhantomBuster, Waste Vision, MaintainX and JCommerce for their feedback that inspired these updates.

Upload files to train Pluno

You can now upload files directly to Pluno so it can learn from your existing documentation. This means you are no longer limited to URLs or public pages.

This makes it easier to onboard Pluno with internal knowledge such as troubleshooting guides, PDFs, specs, or exported docs. Simply upload your files and Pluno will use them as part of its learning sources to deliver more accurate answers and insights.

AI-powered Support QA updates

• Improved QA charts. Graphs now interpolate when data is missing, and time is displayed as hours and minutes instead of large minute values.

• Improved the QA prompt to correctly distinguish public comments from internal notes.

• QA access management is now live. Support agents can view their own QA reports but no longer have access to QA configuration settings.

Other fixes and improvements

• Ticket field filling is now restricted to the fields available in the ticket’s form.
• Improved the call summary prompt to avoid unintended JSON-formatted output.
• We now clearly indicate when learning is limited to public pages only, avoiding confusion when the page count shows 0.
• Fixed an issue where existing Zendesk tags were sometimes removed when Pluno added its own.
• Fixed a sidebar access issue for subdomains with multiple Pluno apps.
• Fixed an issue where email invites failed for addresses containing characters like "+".
• Fixed cases where clustering returned empty results.

Special thanks to our users Netsapiens, Inspectorio, Smartness, Kojo and Budget Bakers for their feedback that inspired these updates.

AI-powered Support QA

Designed to help you measure and improve the performance of your support team based on your own quality standards. You can configure your QA criteria using plain-text descriptions and selecting either binary scoring or a 1–5 scale for each criterion.

Once your criteria are set, the AI automatically reviews support tickets, evaluates them according to your defined standards, and provides clear reasoning behind every score. Results are displayed across three views in the Pluno dashboard: Agent Comparison, Criteria Comparison, and Ticket Insights.

The Agent Comparison view offers a full performance overview for each agent against the selected criteria and includes performance progression graphs to track improvements over time or compare agents with one another. In the Criteria Comparison view, you can analyze how performance against a specific criterion has changed over a given period. In the Ticket Insights view, you can explore individual ticket evaluations, view how each ticket scored against your criteria, and see the AI’s reasoning for its assessment.

To begin, set up your QA criteria in the Pluno QA dashboard.

Enhanced Topic Clustering

You can now guide the analysis by entering a specific query in plain-text, allowing the system to focus on extracting insights such as feature requests or churn-related issues, rather than relying solely on default topic discovery. Plus, now you can also segment insights by Knowledge Hub, making it possible to analyse topics independently across different brands or product areas.

In addition to these new capabilities, overall clustering quality has been boosted for more accurate and meaningful groupings, and the interface has been updated with an improved design for easier navigation and understanding of results. Try it out in the Topics & Trends dashboard.

Other fixes and improvements

• Fixed an issue where internally created tickets were being automatically assigned to the Deflection AI agent but were not being escalated to team or responded to, as the presence of internal notes or team messages correctly prevented the AI from taking action. This resulted in tickets remaining assigned to the AI without the team being aware. With this update, such tickets are now immediately escalated back to the team upon detection, and a recurring fallback check has been implemented to ensure any missed cases are identified and escalated promptly.

Special thanks to our users Smartness and Kojo for their feedback that inspired these updates.

Confluence supported for Knowledge Base

You can now connect your Confluence workspace to serve as a knowledge base for the Answer Copilot. Getting started is simple: just authenticate your account and grant Pluno access to your Confluence instance. Once connected, head to the Pluno dashboard and select the pages you want the AI to use as part of its knowledge base.

More control over Answer Copilot output

If you’ve ever wanted Answer Copilot to generate the exact kind of message you need on the first try, without having to refine it through the chat later, you can now do exactly that. Just click the pen icon on the Find Solution button within the Answer Copilot card in the Zendesk sidebar, then add your custom instructions. Answer Copilot will use those directions right from the start, producing a tailored, ready-to-use response immediately.

Optimised AI Field Filling for more automation with fewer steps

With a single click, you can now automatically enrich any Zendesk ticket: Pluno will populate relevant fields with tags, generate a polished resolution message using Answer Copilot, and even update the ticket title based on the conversation context. To try it out, open the dropdown in the AI Ticket Fields card in the Zendesk sidebar and select this one-click automation option, then simply press the button to trigger it.

Other fixes and improvements

• Now you can have Zendesk tickets tags and fields filled out using AI upon creation automatically without any manual triggers, which was previously limited to deflection-based escalations only.
• Fixed an issue preventing users from inviting team members to the Pluno dashboard due to Slack rate limits. Now new users can be invited using their email only as well instantly.

Special thanks to our users Smartness, Runpod, Almavia CX, and Innovorder for their feedback that inspired these updates.

Zendesk Calls Summary and Insights

Now your teams can quickly understand and act on customer call conversations. Automatically transcribes call audio into clear text and use AI to highlight key points such as pricing, sentiment, and next steps. Easily generate concise summaries to review important topics and identify actionable items without listening back to full recordings.

Setup is simple: connect your Zendesk instance, enable Call Summaries (auto enabled by default), and add your instructions in the Pluno dashboard. Summaries then appear automatically as internal notes inside each ticket.

New Dashboard Flow

We’ve redesigned our dashboard to make navigation more intuitive and user-friendly. The new flow guides different types of users toward the features most relevant to them, without overwhelming them with too many options at once. By simplifying the layout and clarifying the pathways through the app, teams can onboard more quickly and find what they need with less effort. This cleaner structure also makes it easier to fine-tune configurations and get value from the product right away.

Other fixes and improvements

• We now update Jira issue details directly in the linked Zendesk tickets whenever the Jira issue’s project, issue key, or issue type changes. This prevents the broken links seen earlier and ensures all linked Zendesk tickets stay accurate and in sync with Jira.
• Previously, test activity was counted as real usage, causing inaccurate stats. Removing these tests from production stats ensures reporting reflects only genuine customer interactions.
• AI ticket filling is now disabled for closed tickets because Zendesk doesn’t allow field updates on them. Previously, attempts to fill fields on closed tickets caused errors, so this change prevents users from triggering actions Zendesk doesn’t support.

Special thanks to our users Innovorder, Kojo, and Smartness for their feedback that inspired these updates.

Focused Optimization Across all Brands with Separated AI Workflows

AI workflows let you customize how your AI behaves. Previously, creating a workflow changed the AI’s behavior across all brands, which wasn’t ideal and limited your ability to make brand-specific optimizations due to shared settings. Now, you can create separate AI workflows for each brand, enabling precise adjustments tailored to each one.

You can configure AI workflows for every knowledge hub using the dropdown menu in the header or the “Add” button in the knowledge base. Each knowledge hub includes both Deflection and Copilot workflows. However, workflows of the type “for tickets escalated to Jira or Slack” are only available in the default knowledge hub.

You can also copy an existing workflow from another knowledge hub and adapt it using the “Add workflows from another knowledge hub” option.

Other fixes and improvements

• When regenerating AI responses in the dashboard for testing, you can now include the ticket itself as a knowledge source. This lets you see exactly how similar tickets will be handled by the AI going forward.
• The Zendesk standard field “Priority” is now available for AI auto-tagging and categorization.

Special thanks to our users Smartness and Innovorder for their feedback that inspired these updates.

Support Multiple Brands with Dedicated Knowledge Bases

Now you can easily power the support for all your associated brands with Pluno AI. Previously, our dashboard allowed only one knowledge base, so if you linked two different help centers, the AI could get confused. But now, you can create a dedicated knowledge base for each brand within the same dashboard. Just click a button to create a new knowledge base, which will stay separate from others, and then select your brand from the dropdown in settings to link it to the correct knowledge base.

Testing Made Easier

If you want to test the deflection AI before deploying it, you can now easily do so through our dashboard. Testing options are available for both ticket and messaging channels. With one click, send an email to see how Pluno AI responds, or for messaging, add your Zendesk widget key or paste the URL where the widget is deployed to directly simulate the real experience.

Other fixes and improvements

• You can now use Answer Copilot workflows to customize Troubleshooting Copilot behavior as well.
• Previously, sometime an escalation to Jira would fail due to file size checks, but we’ve improved the logic. Now, we directly access files from Zendesk to avoid issues with large attachments during escalations.

Special thanks to our users Smartness, NetSapiens, and BudgetBakers for their feedback that inspired these updates.

Enhanced Capabilities of the Answer Copilot

You can now create separate workflows for the Answer Copilot, allowing you to customize the behavior of both the Deflection Agent and the Answer Copilot independently. This removes the previous limitations caused by the two modules sharing the same workflows.

Separating the logic for these two modules now ensures the Answer Copilot always generates an answer. Previously, it wouldn’t respond when the Deflection Agent classified a ticket as requiring escalation to the team.

As a cherry on top, the Answer Copilot can now also draw information from connected Jira issues, enabling it to provide more informed, context-aware responses.

View all Pluno-handled Chats in Zendesk for a Comprehensive Analysis

Previously, chats handled and resolved by the Pluno AI agent without human intervention were stored only in Pluno’s database and not recorded in Zendesk. You can now enable an option in Pluno’s settings to sync these chats with Zendesk, giving you a complete view of all activity and enabling detailed analysis from a single source.

Other fixes and improvements

• On the Tickets and Chat History page in Pluno’s dashboard, the filter for viewing tickets handled by Pluno previously included cases where the AI only sent a greeting and received no user response. These have now been excluded to reduce unnecessary clutter.
• Updated the UI of the AI Categorization and Tagging page in the Pluno dashboard for greater clarity, including the addition of placeholder descriptions.
• A new “Process Ticket” button now appears in the sidebar when a ticket isn’t found in our system (due to an error or during initial import). This serves as a fallback, allowing you to trigger the import and then handle the ticket with Pluno as usual.

Special thanks to our users SolaX Power, BeatStars, Smartness and Innovorder for their feedback that inspired these updates.

AI Ticket Fields: One-click fill for Tags, Type & Topic

Agents can now auto-populate Zendesk fields with a single click. The button analyzes past tickets plus your team’s instructions to suggest the right tags, type, and topic, speeding up categorization and improving consistency.

Enable it first in onboarding → AI Categorization & Tagging

Clearer Reporting: Escalations split out from AI stats

On the reporting page, escalation metrics now live in their own section, separate from AI performance. This makes it easier to track volume and trends for escalations without mixing them with deflection/AI answer rates.

Topic distribution over time Trend Graph

On the Clustering page, you’ll now always see a topics-over-time graph. It visualizes the immediate children of whatever node you select in the tree, updating dynamically as you navigate, great for spotting spikes or regressions within a specific area.

Smarter Escalations from Messaging: Auto Subject & Requester name

When escalating messaging conversations to Zendesk, Pluno now generates a ticket subject and a human-readable requester name automatically.

You can fine-tune both in the Messaging tab of the deflection config.

Other fixes and improvements

• A reported Answer Copilot bug causing failure in properly ingesting given ticket’s content to generate answers was fixed.
• Auto-reply loops eliminated to detect out-of-office and similar auto-responses and handle them without replying, preventing infinite reply loops.
• Faster Jira escalations by adding caching and optimized AI steps to reduce end-to-end escalation time.
• Personalized ticket examples generation made more resilient by ensuring if a backend deploy interrupts the process, we automatically resume and complete the pipeline afterwards.

Special thanks to our users SolaX Power, Kojo, Smartness, and BudgetBakers for their feedback that inspired these updates.

New Reporting Page for Sharper Analytics

Our new reporting page provides a quick overview of ticket volume and resolution performance, including:

• Total Tickets over the selected period
• AI Resolved and Team Handled counts with percentages
• Internally Escalated tickets
• Ticket Volume by Resolution Type (daily breakdown of AI vs. Team resolutions)
• Resolution Distribution (%) (trends of AI vs. Team resolution share over time)

Required Fields Collection for Deflection Agent-to-Team Escalations

When the Pluno deflection agent escalates a ticket to a team member, it now ensures that all configured required fields are collected before escalation. These fields can be set up in Zendesk and then managed directly in the Pluno dashboard.

Customized Workflows by Channel

Our AI prompts now include the channel type (web widget, WhatsApp, etc.), as well as the current time and weekday. This enables customized AI workflows for specific circumstances. For example, for certain channels, escalations can now require specific information from users before proceeding, ensuring all necessary details are collected first.

Tagging Implemented on Escalation and Resolution

We now automatically add jira_escalated and slack_escalated tags when a ticket is escalated, and a jira_resolved tag when the linked Jira issue is marked done/resolved.

Ingest Non-Public Help Center Articles

Our system can now ingest non-public Help Center articles, allowing AI agents to access internal documentation. This feature is turned off by default and can be enabled as needed.

Personalized Ticket Deflection Mock Example Available for Non-Pluno Users

Non-Pluno users can now preview how ticket deflection could work for their company with Pluno AI through this page. Just enter your company name or domain, and a mock example is generated instantly.

Other fixes and improvements

• Fixed an issue where, when the deflection agent escalated a ticket in Zendesk, only the last 10 messages were carried forward due to a default Zendesk setting. We’ve resolved this by overriding the configuration so the full conversation history is included.
• We now embed documentation content directly for training instead of generating questions for each chunk, making knowledge retrieval more robust.
• Added a filter on the Web Chat History page to show only tickets with Pluno messages.
• Pluno web app admins can now invite users via email directly from the app.
• Fixed an issue where Answer Copilot sometimes got stuck in a continuous loading state.
• Improved AI agent robustness with multiple fallback layers.
• Resolved an issue where Zendesk web chats were not correctly associated with users who had multiple emails linked to an existing account.
• You can now disable Pluno for ticket deflection completely. Previously, once an agent was selected for Pluno to answer as, it always handled their tickets even without an active trigger.
• When using the web widget, end users no longer have to provide their email for escalation if they are authenticated.
• Fixed an issue where the Pluno AI agent continued responding after escalation to a team member.
• Fixed an issue where the channel selector for linking to an existing Jira ticket showed all Jira issues, causing child issues to be mistakenly selected as parents. The drop-down now only displays parent issues.

Special thanks to our users SolaX Power, Innovorder, Kojo, Smartness and BudgetBakers for their feedback that inspired these updates.

Meet the new Deflection Agent in Zendesk

Other deflection agents may have failed you with generic answers, frustrating loops, or even hallucinations. Now, experience a deflection agent fine-tuned to your domain, trained on your space’s context, a full-fledged, advanced knowledge-bearing support agent, just like an in-house trained expert. Deflects queries directly within Zendesk (supports both messaging and tickets), and when it can’t, escalates seamlessly to your team while gathering all the necessary information. Your team also gets full access to Pluno’s dashboard for effortless knowledge base management.

Other fixes and improvements

• Separate workflows can now be configured for the deflection agent and the escalation copilot, allowing each to operate according to your desired behaviors.
• Now you can search for past Zendesk tickets in Pluno’s knowledge base using their IDs.
• To avoid unnecessary inclusion of attachments from a ticket during escalations from support to other teams, we’ve disabled the auto-attach functionality.
• Made the Answer Copilot easier to use by adding a button that inserts the generated answer directly into the editor, ready to be adapted or sent.

Special thanks to our users Netsapiens, Innovorder, Kojo and Smartness for their feedback that inspired these updates.

Introducing Answer Copilot

Answer Copilot is now available in the Zendesk sidebar to help support agents respond faster and more confidently. When a new ticket comes in (pre-escalation), agents can click on ‘Find Solution’ in the Answer Copilot card to instantly generate a suggested reply. Each answer comes with clear reasoning and references, so agents can trust its accuracy before sending. If more detail is needed, agents can chat directly with the copilot to pull the right information from the knowledge base, no extra searching or effort required.

Other fixes and improvements

• Answer copilot now automatically responds in the language of the ticket, instead of defaulting to English.
• Our Escalation agent now uses GPT-5, further improving answer accuracy.
• Fixed an issue where escalation content sometimes failed to appear in the sidebar if it took longer than 30 seconds to load.
• Fixed an issue that prevented custom fields in the escalation form from loading correctly.

Special thanks to our users Netsapiens, Innovorder, Kojo and RunPod for their feedback that inspired these updates.

Unlink Zendesk Tickets from Jira Issues & Slack Threads

Unlinking lets you stop a Zendesk ticket’s updates from automatically appearing in an irrelevant Jira issue or Slack thread and vice versa. It’s also to avoid triggering Jira workflows from irrelevant Zendesk activity, and preventing Slack notifications about outdated support updates. Now teams can manage their work without being disrupted unnecessarily. To unlink, just go to the Zendesk sidebar and click “Unlink” on the linked issue card.

Support Ticket Topic Clustering

Zendesk ticket topic clustering groups similar tickets together based on content, like subject or keywords, so you can spot common issues, trends, or areas needing fixes or improvements. It also helps identify knowledge gaps by revealing frequent questions or problems without clear answers or existing solutions. To view your topic cluster, go to this page of the dashboard, select the time range, and generate the report.

Other fixes and improvements

• Introduced an "Add Files from Ticket" button to attach all ticket files to escalations for forwarding to Jira.
• Fixed the issue with escalation content being lost when sent to Slack. Ensured all content; messages and attachments, are carefully shared.
• Updated misleading icons for certain navigation items in the Pluno web app for clarity.
• Added a Zendesk login link to our login page to ensure easy routing for visitors using different Pluno products.

Special thanks to our users Avi, RunPod, and Smartness for their feedback that inspired these updates.

Improved Jira Modal with User Mentions

We enhanced the modal for linked Jira issue details available in the Zendesk sidebar by adding the ability for support agents to tag a Jira user when adding a comment in the Jira Issue from Zendesk, ensuring they’re notified. To tag, just write ‘@<Jira user name>’ in the comment.

Multiple Slack Channels supported for Escalation

Support agents can now choose different Slack channels for escalating a Zendesk ticket by selecting it from the drop-down menu before escalating. This way, they can ensure the ticket reaches the right team. To have more of your Slack channels available in the drop-down menu, you can add them on the Slack configuration page of our dashboard.

Zendesk Ticket Details Visible in the Jira Issue

Engineering teams can now easily see which Zendesk tickets are linked to an issue directly in the issue sidebar in Jira. This not only makes it easier for them to access relevant information but also sheds light on the severity of an issue. Plus, communicating updates becomes much easier since repeat/duplicate tasks are prevented.

Other fixes and improvements

• Support added for Jira priority field in escalation templates.
• The escalation template now includes the Zendesk username of the agent escalating the ticket for better issue handling.
• Now we display all linked tickets in the root message of Slack threads.
• Clicking on attachments in Zendesk linked Jira issue modal now enlarges images and downloads non-image files.
• Fixed the occasional failing of Jira attachment images to load.
• Added support for .csv file attachments.
• Resolved bugs around multiple Jira instances support.
• Removed unnecessary indicators (on-screen notifications) during Jira issue creation.
• Enhanced attachment forwarding based on AI decisions.
• Increased the robustness and maintainability of the Jira issue creation template.
• Improved the Pluno app behavior inside Zendesk and Jira by making the following changes to the default workflows, instructing the agent to:Check in with the internal team only once after a maximum of 96 hours if awaiting their actionAvoid sending unnecessary messages to Jira/Slack unless the internal team requests informationIgnore changes in the assignment of the Jira issue and do not pass notifications to Zendesk regarding themInform the support agent only if the Jira issue status changes to 'Done' or 'Resolved'Never message about already resolved or closed Zendesk tickets
• Made the text field for commenting dynamic in the linked Jira issue modal, enabling height to adjust automatically.
• You can now submit feedback directly on the Zendesk sidebar without having to send us an email.
• Fixed an issue that was preventing the hybrid agent to distinguish between AI and non-AI messages correctly.
• Optimized high-load database queries, reducing query count by up to 30 times and eliminating timeout errors to ensure uninterrupted service.

Special thanks to our users Innovorder, Embat, AlfredCamera, and RunPod for their feedback that inspired these updates.

Link Multiple Zendesk Tickets to Existing Slack Threads or Jira Issues

With the ability to link multiple Zendesk tickets to an existing Slack thread or a Jira issue, you can track related customer problems in one place, understand the overall impact, collaborate more effectively, and resolve the root issue faster. At the same time, it helps reduce duplicate work and provides a clearer audit trail for incident reviews or reporting.

Other fixes and improvements

• Agent's name is now included in comments made from Zendesk to Jira. (based on feedback from Embat)
• Added the ability to find and link Jira issues using the issue link.
• Resolved issue where multiple repetitive comments were sent to a Jira issue when several Zendesk tickets were linked.
• Reduced the information transferred from Zendesk to only essential fields for processing ticket events. (based on feedback from RunPod)
• Fixed an issue where agents, who didn't install the bot on the Zendesk subdomain, couldn't see the Jira issue modal. (reported by ScopeAR)
• Fixed an issue where long conversation messages, like code or encoded text, weren't processed correctly due to context window limits in our AI models.
• Now display the original Jira error message if escalation to Jira fails, aiding users with Jira issue validator workflows. (based on feedback from Innovorder)
• Fixed the handling of component fields during Jira escalation.
• Improved escalation prompts and template filling to correctly input company names, reducing the need for manual changes.
• Fixed an issue where images in the Jira description were not displaying correctly in the Zendesk popup. (reported by Embat)
• Improved onboarding flow:Status is now inferred from existing data, preventing discrepancies where completed steps don't appear on the onboarding page.Added a warning for users who haven't updated their Jira escalation templates and a message informing users that we support multiple Jira instances after installation.Renamed "Zendesk Web Widget Authorization" to "Zendesk Messaging Authorization" for clarity
• For communities with the Zendesk app installed but incomplete onboarding, we now show a message indicating the need to finish onboarding to escalate to other platforms. (based on feedback from Solax)
• Resolved a lingering issue with attachments that persisted after the previous fix.
• Added a 404 page for non-existent or moved pages.
• Implemented a loading skeleton to show while determining the community onboarding state.
• The text field in the web tickets inbox now automatically expands to a set height as you type. (based on feedback from Best Wallet)
• Auto-escalation with the hybrid agent now functions for private inbox channels on Telegram.

Special thanks to our users Innovorder, Embat, Best Wallet, Solax, RunPod, and Smartness for their feedback that inspired these updates.

Jira Comments and Attachments in Zendesk

Pluno now display Jira issue comments and attachments directly within Zendesk, enhancing visibility and context for support agents. Additionally, agents can now send comments to Jira issues straight from Zendesk.

Support for Multiple Jira Instances

You can now connect additional Jira instances to easily manage tickets across multiple departments, brands, or workflows. Once configured, you will be able to select which Jira instance to escalate issues to directly from the Zendesk app.

Linked Jira Issue Card in Zendesk App

The Linked Jira issue card is now present in the Zendesk sidebar, providing detailed information about the any associated Jira issues. This will prevent any duplicate issue creations, plus provide support agents with context and information needed to handle an existing or recurring issue.

Other fixes and improvements

• Adjusted the default follow-up time with the internal team from 6 hours to 96 hours.
• Removed the default step of adding a summary in the internal chat when a ticket is resolved.
• Updated default Jira and Slack templates to include an instruction to "fill out the template in English," clarifying that you can change the language.
• Support added for Zendesk placeholders in web widget-created tickets, retrieving values directly from Zendesk for more robust escalation content.
• Customers can now complete onboarding steps 2 (API key) and 3 (Zendesk messaging) at any time.
• Fixed an issue where the escalation template was occasionally not followed.
• Fixed an inconsistency with the onboarding status, ensuring everything is up-to-date after the database migration.

Special thanks to our users Embat, Corsair, BeatStars, and Cloudhuset for their feedback that inspired these updates.

Attach files to escalations from Zendesk To Jira

Upon the escalation of a ticket to Jira, files can be attached to the Jira issue directly from Zendesk. Any files that are shared on Jira are automatically being forwarded to Zendesk.

Support for login with Google

Now you can log into Pluno with your Google account as an alternative to email.

Enhance escalations with Zendesk placeholders

Pluno automatically uses your pre-configured Zendesk placeholders to fill in escalation templates to Jira and Slack. You can also insert these placeholders manually in templates you set up in our web app, for example: {{ticket.id}}.

The default Jira template now includes the Zendesk ticket id and link.

Summaries in agent language

The ticket summary is now in the support agent's language based on the previous messages in the ticket. Thanks to Irene from Embat for the feedback!

Other fixes and improvements

• When switching project or issue type, all filled non-custom fields will retain their values. Thanks to Irene from Embat for the feedback!
• If required fields are empty during escalations, the escalate button stays disabled and highlights what’s missing.
• All required Jira fields are now always shown during escalation.
• Dependent dropdown Jira fields are now supported.
• The hybrid agent no longer posts an internal comment for the escalation itself.
• For subtasks we now show a default option for the parent during escalation.
• You can now invite the Jira admin during onboarding, so they can install the Jira app.
• The Zendesk sidebar now shows an hint if onboarding is not complete.
• You can now configure the default Jira project in the app configuration on Jira.
• The Zendesk subdomain field during onboarding now clearly shows which part of the subdomain is needed.
• If you haven’t completed the Zendesk onboarding and want to continue, you no longer have to reconnect your Zendesk account again.

Change of subprocessors

We've introduced PostHog as a subprocessor to improve resolution time of bugs & reported issues, as well as gaining insights for product improvements.

The Pluno Copilot App is Now Available on the Zendesk App Marketplace

You can now find Pluno Copilot on the Zendesk App Marketplace!

The app helps you escalate by sending tickets from Zendesk to Jira or Slack with a single click. Pluno gathers all the relevant context; ticket details, conversation history, and troubleshooting steps into a clean summary using a customizable template, so that the receiving team can jump in quickly. If both Slack and Jira are connected, you can choose where to escalate each ticket.

Once escalated, Pluno then keeps everything in sync by:

• Adding updates from Jira or Slack as internal comments in Zendesk
• Generating suggested messages to help agents respond to customers and keep them updated with just a few clicks.
• Syncing new and relevant customer messages back to the original escalation thread

This means agents can stay in Zendesk while collaborating with other teams and keeping customers up to date - without manually tracking multiple apps or chasing updates.

How to Set Up the Integration

Install the Pluno Copilot App in Zendesk

Go to: https://app.pluno.ai/zendesk/signup

Zendesk Integration - Fixes and Improvements:

• When escalating a Zendesk ticket to Jira, all Jira fields - including custom fields, are now supported. Users can configure which fields appear directly from the Zendesk sidebar using the “configure fields” button.
• We now support Web Forms as an incoming channel in Zendesk. End users can submit tickets directly through the Help Center on their Zendesk subdomain, providing more flexibility in how they create support requests.
• When escalating a Zendesk ticket to Jira via the Hybrid Agent, file attachments are now supported. You can send .txt, .png, .jpeg, .gif, .pdf, .doc, .docx, and .mp4 files from our system to Jira. Files sent from Jira back to our system support the same formats, except for .mp4, which is not yet supported.
• Pluno imports and learns from the last 100 tickets during installation.
• The first suggested message shown after an escalation is now AI-generated. It adapts to the style of the conversation. This replaces the fixed message previously used with one that feels more personalized.
• The onboarding overview page now clearly lists the conditions required to complete the Zendesk setup.
• Internal comments in Zendesk now show “Pluno:“ instead of “Pluno - Update from Jira:” when the update isn’t actually coming from Jira. This prevents misleading labels on messages triggered by support agents or other non-Jira sources.
• Zendesk ticket status changes are now synced into our system. When the status of a ticket is updated in Zendesk, the new status is also shown in the corresponding Slack thread.
• We now handle changes of the Zendesk subdomain.
• Changes in Jira, like assigning an issue or updating its status, are now monitored by our system. This allows the hybrid agent to detect those changes and post an update in Zendesk.
• Resolved inconsistent loading behavior across the product. In some cases, loading indicators would run indefinitely. These have now been corrected.
• Fixed an issue where Zendesk ticket status was not handled properly for non-English instances.
• Zendesk tickets no longer switch to “Open” when Pluno adds an internal comment from an agent or end user. However, if the internal comment is triggered by an update from Jira or Slack, the ticket status will switch to “Open”.

Other fixes and improvements

• The web chatbot now supports the ‘allowNewMessageNotificationBadge’ property. This lets users choose whether to show or hide the red notification badge when integrating the chatbot into their website.
• You can now create, view, and edit escalation templates from the Escalation Templates page in the Pluno dashboard. Templates now support advanced field types including numbers, strings, and arrays, and come with improved default values. They work across all integrations - Slack, Jira, and Zendesk, and are now included in the onboarding flow to ensure they’re ready to use from day one.
• Users no longer need to manually add our Slack bot to individual workspace channels. This step is now included as part of the installation flow, making setup easier and more streamlined. You can try it out by installing the bot using this link.
• We’ve improved the Hybrid AI’s internal comment behavior by refining its prompt. It now avoids repeating content from the user’s message and clearly suggests when a reminder should be set, instead of setting it automatically without notice.
• When adding the Pluno app to Slack, users choose a channel where escalated tickets will be sent. In large workspaces, the list of channels can take time to load. We’ve added a clear loading message so users understand what’s happening and aren’t left waiting without context.
• Slack messages that include lists are now displayed correctly.

Greater Control and Flexibility for Support Teams to Switch From Automated to Human-Led Mode

Support teams have more control over when to switch into human-led mode — either by replying directly from the web ticket inbox or by manually triggering the handoff from Slack or Discord using the!takeovercommand. This allows you to step in when a ticket needs to be handled by a human.

Zendesk Integration - Enhanced Hybrid AI Behaviour

We’ve introduced enhancements to the behaviour of our hybrid AI agent in the Zendesk app to support a smoother workflow and clearer communication for you.

Improved AI Suggestion Logic

The Pluno Hybrid Agent now holds off on generating a suggested answer when there isn’t enough information to produce a meaningful response. Instead of removing the suggestion entirely, the app now displays a clear message so agents understand why no suggestion is shown. To support this, we have introduced two distinct states for AI suggestions:

1. Not Enough Information - the AI doesn’t have the context it needs to suggest a response.
2. No Suggestion Needed - shown after an agent sends a message or manually discards a suggestion.

Unified Language in Messaging

After an escalation (i.e. once a Slack thread has been created), the AI refers to the internal team as the “support team,” providing a clear and consistent way for users to understand who is handling their issue. Previously, the AI referred to them simply as “team,” which lacked clarity in identifying the specific group involved.

Loading Indicators

Loading indicators are now shown in more places throughout the app—including when the AI is generating a suggestion and when the “Escalate to Slack” button is clicked. These visual cues give you immediate feedback that an action is in progress.

Suggested Answers Now Auto-Hide After Use

Once a support agent sends a message after viewing a suggested answer, the suggestion is automatically hidden. This ensures the interface stays focused and free of outdated content, whether the suggestion was used or not.

Zendesk Integration Fixes

This release includes a set of refinements to the Zendesk app integration, focused on improving usability, visual consistency, and internal handling for you.

• Zendesk knowledge search now omits common stopwords (e.g., “a,” “I”) from highlighted results. This makes search results easier to scan and more relevant to what agents are actually looking for.
• Suggested answers in the Zendesk integration no longer recommend reaching out to the support team. Instead, suggested answers now include meaningful responses based on the user’s query, making them more useful and actionable for agents.
• Frequently Asked Questions now appear fully expanded in the search results—no need to click to view the complete answer.
• The “Escalate to Slack” button now remains visible even if the ticket summary is still loading.
• Messages written to the end user in the Zendesk app are no longer forwarded to the internal Slack channel. This change ensures that only system-generated updates and escalations appear internally, helping reduce noise.
• The “Insert into Editor” and “Discard” buttons in the Suggested Message component are now fully functional. Selecting “Insert into Editor” no longer causes the suggestion to disappear.
• The component previously called Update from the Team has been renamed to Suggested Message to better reflect its purpose. In addition, the label shown to agents now changes depending on the stage of the support conversation:Pre-escalation: shown as Suggested Answer — a potential response the agent might send to the user.Post-escalation: shown as Suggested Message — a draft message for the agent to review, edit, and send if needed.
• Internal notes created in Zendesk are now processed and stored in our system, allowing the hybrid agent to use them as context when considering the next steps or generating a new message suggestion.
• The hybrid agent now clearly labels its suggestions as notes for support agents, rather than implying they are messages being sent to users. This helps agents understand that the content is a draft for their review—not an automated response.
• Agents can now use AI to tailor suggested messages after an escalation. Previously, this capability was only available during the pre-escalation stage. This update brings consistency across both stages of support and gives agents greater flexibility to refine their responses before sending them to users.
• Suggested answers in Zendesk now include the subject line of the user’s email. This provides agents with more context and helps ensure the AI-generated response aligns with the original inquiry.

Other fixes and improvements

• End users can now upload Word documents (.doc, .docx) across all supported platforms with a size limit of up to 10MB.
• You can now copy the team invite link anytime using the new “Copy Invite Link” button on the Members page. This makes it quicker to invite others without needing to start a new invite or ask an admin for help.

Pluno Now Learns From Past Zendesk Tickets! 🎉

We now import past Zendesk tickets into our system during installation. This ensures Pluno can learn from previous customer interactions, leading to more accurate responses while also preserving ticket history, so valuable context from past conversations isn’t lost.

File Attachments Supported in Zendesk

Pluno’s Zendesk Integration now supports file attachments!

• End users can send file attachments directly through Zendesk.
• For escalated tickets, you can send files from Zendesk to the internal channel.
• Files can be forwarded from internal channels to Zendesk as an internal note via the Pluno Hybrid Agent.

This makes it easier to share important files, such as screenshots and documents, enhancing communication and collaboration between teams and the end users.

Pluno and Zendesk Installation Improved!

Previously, installing Pluno required manually creating a webhook in Zendesk to enable the web widget integration. Now, this webhook is automatically generated during the installation process. This reduces manual work, speeds up setup, and ensures a smoother integration experience.

Other fixes and improvements

• A bug has been fixed where code sent through the web widget was displaying incorrectly . Now markdown code is rendered correctly.
• When tickets are escalated from Zendesk to Slack, threads will now display the customer’s name from the conversation widget instead of “Zendesk User”.
• Suggested answers from Pluno in Zendesk now retain their original formatting, including proper spacing and line breaks.

Zendesk Integration with Pluno Hybrid Agent - Now Live!

Instantly escalate tickets to the right team on Slack along with a Pluno generated ticket summary. Teams get a clear, ready-to-use overview, so they can act fast without back-and-forth.

• Pluno handles relevant updates and messages between internal teams and end users, ensuring everyone stays informed without extra manual effort.

• AI-powered suggested responses allow you to speed up your response time.
• The entire knowledge base is at your finger tips. Search for related content from past resolved tickets, web pages and FAQ’s all within Zendesk.

By working directly within Zendesk, Pluno helps support teams respond faster and manage tickets more efficiently. Customers get timely, consistent updates, and agents spend less time on repetitive tasks - all within the platform they already use

We Have Achieved SOC 2 Type II Certification!

We are excited to announce that we have successfully completed the SOC 2 Type II audit. This milestone underscores our commitment to securing our customers’ critical information assets and maintaining the highest standards for data security, confidentiality, and privacy.

This achievement highlights our dedication to protecting sensitive information and ensuring trust with our customers. We will continue to uphold these rigorous standards to provide a secure and reliable service.

Design Update

The redesign of the Web Widget and Ticket Inbox has now been fully deployed, providing a cleaner, more intuitive interface for a better user experience.

Other fixes and improvements

• Typing in the web widget on iOS mobile devices no longer causes unintended zooming.
• Fixed an issue where messages from support agents requiring translation would display as “translating…” until the page was reloaded. Now, translations appear instantly without needing to refresh the page.
• Workflow names and descriptions now update correctly after saving, instead of reverting to the previous version.
• For the announcement translation feature on Discord, channels without the necessary permissions now appear unselectable in the dropdown menu, making it clear which channels cannot be chosen.
• Made improvements to the Knowledge Hub so that content no longer disappears while editing.

Our Member Access Dashboard is live!

Community admins can now manage team access to the Pluno web app by which includes inviting new members, assigning roles (Admin or Support), and removing dashboard access when needed. When an admin sends an invite to their team member, the user will receive a DM on Discord or Slack from the Pluno AI Agent with a link to join, making it easy to onboard and manage team members.

Knowledge Search

We have launched knowledge search! Unlike traditional keyword searches, this contextual search understands the meaning behind your query rather than just looking for exact word matches.

File Attachments Now Supported for Escalated Tickets in Hybrid Mode

Users can now attach files when escalating a ticket via the Hybrid AI Agent, making it easier to share relevant context and resources. The AI Agent will facilitate file sharing between the end user and support agent. Each message supports up to 10 files with a maximum size of 10MB per file, allowing flexibility in sharing text files, PNG, MP4, GIF, and JPEG formats. This update improves the escalation process by enabling users to provide screenshots, logs, or videos for faster issue resolution.

Improved Ticket Search

We have now made it easier to search for conversations in your ticket inbox with our new search box. You can now quickly locate tickets based on:

• Specific words or phrases
• Email or Customer ID
• Conversation ID

This update helps you navigate your inbox more efficiently and get to the right conversations faster.

AI Agent now replies with FAQ answers for keyword matches

The AI Agent now responds with FAQ answers when a end user message matches a FAQ entry. Previously, it would only reply if it detected a question or problem. Now, if someone types a keyword or phrase that matches a predefined FAQ entry, the AI Agent will automatically respond with the corresponding answer.

This makes it easier for users to get relevant information without needing to ask a full question.

Other fixes and improvements

• Resolved an issue where role permissions were not being applied correctly on the roles page.
• Updated how renamed Discord channels are handled. When the bot references a knowledge item that mentions an old channel name, it will now correctly point to the updated channel to prevent outdated information.
• We have implemented further security measures, including enhancements to the invitation system for the member dashboard. Now, only explicitly invited users can join, ensuring stricter access control and improved protection of sensitive information.
• Fixed a bug that caused the ticket inbox to reload indefinitely.

Launch Week - Day 5

We’re excited to introduce Pluno’s Post-Escalation Workflows, designed to make handling escalated tickets smoother, faster, and more aligned across teams.

Just like Pre-Escalation Workflows, setting these up is quick and simple. No complex setup—just describe the conditions in plain language, and Pluno’s AI will automatically identify the right triggers based on intent, urgency, and context.

Feature 5️⃣ : Post - Escalation Workflows

These workflows can help automate follow-ups, keep teams aligned, and eliminate manual status tracking. For example, they can ensure a refund is confirmed before notifying the customer or verify that Engineering has fully deployed a fix before an update is sent. These are just a few ways Pluno streamlines the escalation process - workflows can be customized to fit your team’s unique needs.

To get started, visit the Pluno AI Workflow dashboard (Settings → AI Workflows).

See it in action—watch the video above!

Other fixes and improvements

• Suggested questions and answers on the web widget have now been included in the hybrid agent’s initial message.
• Resolved an issue in Slack where the agent assignment message did not include whether the ticket was hybrid or human-led.
• The Hybrid Agent now processes multiple messages without replying to each one individually. Whether from a customer or a team member, it processes the full conversation before responding.

Launch Week - Day 4

Feature 4️⃣ : Reporting Page

The new Reporting Page is now live, offering a more structured and intuitive way to track key metrics and performance trends. This dedicated space provides deeper insights into how the AI Agent is performing, including details on tickets resolved, resolution rate, and response time. It also measures the team’s performance in terms of response time and CSAT (Customer Satisfaction Score), giving a comprehensive view of both automated and human-assisted interactions. With clearer visibility into key data points, it’s now easier to measure customer support success over time and identify critical areas for improvement.To get started, visit the Pluno AI Workflow dashboard (Settings → AI Workflows).

Other fixes and improvements

• Clicking the “Contact” button in private channels no longer shows the default system text. Now, only the configured first message for live support is displayed.
• The welcome message content (e.g., “What is your community name?”) is no longer included in the ticket summary when a new ticket is created—only messages before escalation are shown for clarity.
• The root message for a ticket in Slack now correctly displays the channel name instead of the community server name.
• The first time a user interacts with the hybrid agent, a message appears in the internal channel to provide context. In private channels, this message is temporary and only visible to that user.

Launch Week - Day 3

We’re excited to introduce Pluno’s Pre-Escalation Workflows!

Thanks to natural language processing, support teams can now build pre-escalation workflows without complicated rules or manual setup. Just describe the escalation conditions in plain language, and the system will automatically interpret the right triggers based on intent, urgency, and context.

Feature 3️⃣ : Pre-Escalation Workflows

From gathering customer information upfront, to identifying the steps a customer has already taken to resolve the issue, Pluno’s AI workflows adapt to a variety of use cases, helping teams reduce back-and-forth and speed up resolutions.

To set up these workflows, just go to Pluno’s AI Workflow dashboard (Settings → AI Workflows).

To quickly see it in action, check out the video above.

Other fixes and improvements

• Clicking the “Contact” button in private channels no longer shows the default system text. Now, only the configured first message for live support is displayed.
• The welcome message content (e.g., “What is your community name?”) is no longer included in the ticket summary when a new ticket is created—only messages before escalation are shown for clarity.
• The root message for a ticket in Slack now correctly displays the channel name instead of the community server name.
• The first time a user interacts with the hybrid agent, a message appears in the internal channel to provide context. In private channels, this message is temporary and only visible to that user.

Launch Week - Day 2

We’re back with another exciting update to transform the way you work. Each day of Launch Week brings a new feature designed to make your customer support workflow smoother, faster, and more efficient. Today’s update tackles one of the most common pain points for teams working across multiple Slack workspaces.

Feature 2️⃣ : Cross-Workspace Messaging for Slack

Support agents can now send messages across different Slack servers through the Pluno, which enables the exchange of messages between connected workspaces. This enables seamless communication between teams and external partners without the need to switch between workspaces. As a result, agents can respond more quickly and collaborate efficiently, all while keeping conversations within a single platform.

Other fixes and improvements

• You can now attach video (.mp4) and GIF files, with a size limit of 10 MB. While Telegram functionality is still in progress, full support will be available soon.
• We resolved an issue where hybrid agent reminders and delayed CSAT workflows were disrupted by backend updates. Workflows like resolving after 2 days now work as expected.
• An issue has been fixed where text would appear behind the chat widget in Brave and Firefox browsers.
• Toast notifications have been improved with a new design: positioned in the bottom left, no close button, and a sliding animation that makes them disappear automatically after ~6 seconds.
• A bug was fixed where the summary was not displayed in the web ticket inbox for one day.
• We’ve addressed an issue in Firefox where the “Generating answer…” message would remain visible after the answer was provided.

Launch Week - Day 1

Welcome to Pluno’s first-ever Launch Week! We’re thrilled to kick off a week full of exciting updates, new features, and enhancements designed to make your customer support workflows smarter, faster, and more efficient. Each day, we’ll unveil a new feature along with improvements to help you get even more value from Pluno. Today marks the first of five major announcements, so stay tuned as we continue to redefine what’s possible with support workflow automation.

Feature 1️⃣ : Automatic AI Ticket Assignments and Priority

Tickets are now automatically assigned to the most appropriate groups based on group names and descriptions. This streamlines the categorization process, reducing the need for manual sorting and enabling faster resolution times.

Automated Priority Levels: Ticket priority is now determined automatically by analyzing the user’s query. Distinct symbols are used to represent each priority level: • ❕ Low Priority

• ❗ Medium Priority

• ‼️ High Priority

This update ensures that critical issues receive attention more quickly, while less urgent matters are appropriately queued. The system continuously improves as more tickets are processed, leading to increasingly accurate assignments and prioritizations over time.

Note: Priority levels will become more stable and accurate as more tickets are processed.

Check out the video above to see it in action!

Other fixes and improvements

• Fixed an issue where the CSAT message was missing in web widget integrations. Now, users can easily reply and continue conversations directly from their emails. We’ve also made sure that older versions of the web widget still work properly.
• Improved agent ticket inbox: fixed inconsistent ticket states, corrected lists in web integration, added reply button and email integration for CSAT emails, and adjusted chatbox size to fit content (2-6 rows)
• Chat widget title and first message can now be customized on the chat widget test page. Changes require a page refresh to take effect.
• Fixed a bug in the hybrid agent that caused duplicate messages when handling multiple simultaneous incoming messages.
• Automated AI answers now reflect the custom label for the contact support button, allowing the button text to be customized through the community config instead of defaulting to “contact team”.
• Groups are now live on production! This new page on the dashboard allows you to assign support team members to specific groups, supporting the Hybrid Agent in automatically assigning escalated tickets to the correct group.

Introducing Pluno's Hybrid AI Agent!

We’re excited to announce the release of our Hybrid AI Agent, a new solution designed to streamline and enhance your customer support operations by combining intelligent automation with human expertise. Heres how:

• AI Workflows: Intuitively handles user intent, triggering the right actions without forms or buttons for a seamless experience.
• Smart Ticket Routing: Assigns complex issues to the right team members with a ticket summary including customer details, interaction history, and key insights for faster resolution.
• Automated Customer Updates: Keeps customers informed with progress notifications and resolutions automatically.

Pluno’s Hybrid AI Agent is designed to save you time and allow your team to focus on high-value problem-solving.

Check out the demo video above to see it in action!

AwesomeQA is now Pluno!

Over the past year, we’ve made significant advancements to our platform, and to better reflect our growth, we’ve rebranded from AwesomeQA to Pluno.

Here’s what this means for you:

• Expect the same dashboard and bot functionality. The only difference - a new name!
• All existing links will remain functional. You’ll be automatically redirected from awesomeqa.xyz to pluno.ai.

Now Supporting File Attachments!

AwesomeQA now supports file attachments (PDFs, TXTs, and images) on the following platforms:

• Chat Widget
• Telegram
• Discord
• Slack
• Email

Handling of Unsupported File Types

If unsupported file types are sent via external chats (e.g. Telegram), a security warning will be sent to your dashboard and internal channel with a downloadable link to the file.

Improvement to Email Attachments

Previously, email attachments from users were not displayed in internal chats. This issue has been resolved, ensuring all email attachments are now visible in internal communications.

This feature is now live across all supported platforms. If sharing files has been a missing piece in your workflow, you can start using it today to make issue resolution more straightforward.

Our Bot can now learn from Past Ticket Conversations

Our bot can now learn from past ticket conversations to enhance its knowledge base and provide more accurate and relevant responses.

• Data Privacy First: All confidential information is anonymized to ensure that there is no risk of sensitive information being exposed.

• Improved Responsiveness: This feature provides the bot with additional resources, allowing it to respond to more questions effectively.

Note: Currently, this new knowledge is utilized exclusively for generating responses to tickets, not for answering community questions.

Other fixes and improvements

• Configurable hand-off message, previously limited to web-based escalations, are now available on all platforms. A hand-off message is triggered for every escalated ticket, ensuring consistency across all external platforms.

• A bug on Telegram has been fixed where “Contact Team” button, below previous AI responses in external channels were not removed after a new AI response appeared.
• The redundant title message “Ticket from” has been removed on Slack, streamlining ticket displays for a cleaner and more concise interface.
• The label of the “Contact Support” button has been corrected back to Contact Team.
• Fixed an email spoofing vulnerability in our domain “pluno.ai”.

Now Supporting Slack for Customer Support!

AwesomeQA now supports Slack as a unified inbox alongside Discord! Manage customer tickets from your website, and support emails, and bring all the communication from your hundreds of Telegram groups directly into Slack for seamless management and organization.

Support agents can collaborate with engineers or product teams inside Slack, eliminating the need to switch platforms and reducing miscommunication.

Try it out by adding AwesomeQA to your Slack workspace: Get Started.

End Users Can Now Reply to Tickets Directly via Email

Previously, when an end user received a response to their ticket via email—because they were no longer active on the web widget—they had to click a link in the email and return to the widget to reply. Many users instead replied directly to the email, but those responses were not captured in the system.

Now, end users can reply to the email directly, and their responses will automatically be added to the ticket. This ensures no communication is missed, and users can engage with support seamlessly via email or the web widget.

Other fixes and improvements

• We now support CCs in our email integration! Any email address included in the CC of an email thread will automatically stay in the CC when you send a reply.
• We’ve optimized the web widget for faster loading speeds.
• CSAT emails now include direct rating buttons and display a "thank-you" page after the end user submits their rating.
• To improve accuracy, we now ensure CSAT emails are sent only once—10 minutes after the final resolution—even if the ticket is reopened and resolved again.
• CSAT emails are now sent to end users only when there is an interaction with the team.
• End user email replies in internal communication tool tickets now exclude quoted email content, providing a cleaner and more focused view.
• End users can now reply to ticket emails from any email address, even if they’ve set up email forwarding, and their responses will still be added to the ticket as if sent from the original address.
• Fixed a bug on the settings page where the read/write toggle states were incorrect when multiple Telegram groups were configured.

Integration for Email Support

Managing support inquiries via email is now easier! Instead of having your team manage support emails separately in different inboxes, user emails are automatically converted into tickets and sent directly to a designated channel in your Discord server. This allows your team to respond to and manage all tickets in one place, in real-time, improving efficiency and response times. Your support agents can handle email inquiries alongside tickets from other platforms, like your web chatbot, without needing to switch between tools. To get this set up, please let us know and we’ll help you get started.

Redesigned Web App for a Modernized User Experience

We’ve launched a major redesign of our web app, featuring a fresh, modern look that improves navigation and usability. The new interface is cleaner, more intuitive, and provides a consistent experience. Whether you’re checking analytics, managing bot settings, or reviewing knowledge items, the updated dashboard makes everything easier and more enjoyable. This design upgrade is focused on enhancing productivity and user satisfaction.

Other fixes and improvements

• The web widget now retains its open/closed state when navigating or reloading the page, but only after the ticket is escalated. Previously, it would reopen even after being closed.
• The chatbot disclaimer has been updated to include acceptance of cookies.

A New Level of Multilingual Support

End-users can now send messages in their own language through the web chatbot, which will appear in real time in English on the ticket inbox for your support agents to easily understand. Similarly, your team can respond in their own language, with replies translated into the user’s language. This allows both the AI agent and your team to offer multilingual support in real-time without additional tools. This feature is available for all web chatbot users.

Another Step Towards Data Security

As we work toward SOC 2 and GDPR certification, we are ensuring compliance with their standards. A new disclaimer will notify your web chatbot’s end-users that, by using the chat, they agree to our privacy policy and terms.

Other fixes and improvements

• Refined the AI to improve language consistency and minimize inaccuracies when sources are unavailable.

• Previously, FAQ items added by trusted users were saved for future reference only. Now, the original Discord message is also updated instantly.
• Users can now also sign up/log in via email, making the tool accessible to non-Discord users who want to use it for their websites or Telegram groups.

Assess Your Customer Service Quality With CSAT

Now your users can rate their experience after receiving support from human agents through the web chatbot. You can view these scores on your Web Chatbot Analytics page of the dashboard, and with a time-period-based view, you can quickly understand the quality trends of your service. This allows you to make necessary changes at the right time and ensure that no improvement opportunity is missed.

Handle Website Chats More Efficiently With Macros

If your users create tickets to switch from AI-powered support via the web chatbot to direct support from your human agents, and they still ask repeated or basic queries that are already available in your knowledge base, you no longer need to manually answer these queries. Instead, you can easily create macros that allow you to perform a quick search within the ticket itself.With just one click, the pre-written answers saved in your macros will appear, ready to be sent to the user. This helps your agents save time and focus on more pressing matters. You can easily create macros on your dashboard's designated page, and once created, quickly search using '/'.

Customize FAQ Prompts

Last time, we informed you that we had added pre-defined FAQ prompts to the web chatbot to help your users easily start engaging with the chatbot and give them some guidance on how to navigate your product. These FAQ prompts were automatically configured for you, but after receiving feedback, we realized that you might prefer to configure these FAQs yourself. Now, you can easily set them up according to your preferences on your dashboard's designated page.

Change of subprocessors

To improve response times, we now also send some of the OpenAI requests to deployments on Azure. Please note that this introduces Microsoft Azure as a new subprocessor of your data.

Predefined FAQ Prompts Are Now Available In The Web Chatbot!

Now, when your users open the web chatbot, they’ll see a set of commonly asked questions about your project right away. This makes it easier and faster for your users to find the information they need without having to type out their questions. It’s designed to save time and streamline their experience, especially if they’re new to the platform and don’t know where to start.

This feature is automatically enabled for all users who have their project documents in our system. We’ve pre-configured 3 questions for each project, based on their documentation, ensuring that common FAQs are covered from the start. If any adjustments are needed, our team will handle them directly.

Shareable Ticket Links for Better Collaboration

We’ve introduced a new feature that makes it easy to share individual ticket links. If you’re discussing, for example, a technical issue and want your engineers or team members to review the ticket or take over the conversation, you can now quickly copy and share the link internally. This streamlines collaboration and ensures everyone has direct access to the necessary details. No more searching for tickets—just copy the link with a click and share it instantly! Try it out now from your dashboard.

New Message Classification Approach Goes Live!

Our new approach for classifying messages is now live! Based on the results of our recent A/B test, we’re confident this update will significantly boost our response ratio. It’s also designed to reduce responses to messages that don't necessarily require one, such as general complaints. This improvement will help streamline our interactions and ensure that responses are more relevant and efficient.

Other fixes and improvements

• We now have a status page to show whether our tool is currently up and running. You can check this page for real-time updates on system status.
• Fixed an issue where FAQs created on our dashboard by users weren’t immediately updated.
• Support agents can now easily copy the end-user's email in the tickets management portal by either clicking on the email itself or using the button next to it. This small improvement makes handling user info quicker and more efficient.
• Fixed the FAQ channel detection issue. Previously, if non-trusted user messages were detected, the channel was classified as a non-FAQ channel. We’ve updated this logic: now, if all messages in a channel are from the same author, it’s more accurately recognized as an FAQ channel.
• End users can now have only one open ticket at a time. If they attempt to open another ticket or reply to a resolved one, they'll receive an email with a link to their current open ticket. This helps keep ticket management streamlined and ensures support agents don’t get distracted or lose time due to unnecessary ticket creations for the same issue by the same user.
• Ticket status is now determined based on the latest message in the chat when a support agent claims the ticket, rather than automatically changing to the OPEN status. This change ensures a more accurate status is set, reflecting the most recent activity and improving overall ticket management.

Web Chatbot Disclaimers

Now you can add a disclaimer inside your web chatbot to manage user expectations and reduce liability. Previously this functionality was limited to Discord and Telegram bots only. This is a great way to clarify that the bot is not a human expert, so users shouldn't rely solely on its responses for critical decisions. To set it up, in the chatbot website configuration add your disclaimer in the following setting: “disclaimer?: string; // Add here”.

For reference: Developer Manual.

New Approach for Question Classification

We have launched a new method for classifying messages to identify whether they are questions, and it's currently undergoing A/B testing. This crucial enhancement is designed to improve our response rate to questions while minimizing replies to irrelevant messages. Once the desired results are achieved in production, we will integrate this new solution into our system shortly after.

Other fixes and improvements

• Removed the "Thanks, that's helpful 👍" quick feedback button from the ticket creation process in the web chatbot for the end user. The button is now only displayed when the last message in the conversation is from a human agent.
• Resolved an issue where the bot would stop functioning in Telegram groups after the group ID changed (e.g. when a group transitioned from private to public). The bot now correctly handles group ID changes and continues to operate properly.
• Added a notification message in the end user interface of the web chatbot to indicate when a ticket has been resolved by a human agent.
• Now we send an automated initial message when a ticket is created in the web chatbot based on feedback from GamerGains to improve the user experience.
• Fixed a pagination issue in the Chatbot Support page on the dashboard on large screens, ensuring a consistent and smooth user experience across all screen sizes.
• Resolved layout issues with the web chatbot on smaller mobile screens for improved end user experience.
• The knowledge ingestion process from forum-style FAQ channels in servers now works similarly to announcement channels, to ensure comprehensive storage of all knowledge items.
• Fixed an issue where the web chatbot would frequently repeat information from previously sent messages in a given chat.
• Resolved an issue where the web chatbot recommended users to open a ticket in the Discord channel since tickets can be directly created through the web chatbot itself.
• Announcement messages are now automatically deleted from the knowledge base when the original announcement is removed from the Discord server.
• Improved the web chatbot to correctly direct users to create tickets with their staking providers when their queries are related to staking issues, following feedback from the Lava Network that users were mistakenly being advised to contact them instead.
• Fixed the chat disconnection issue during live chats between a human agent and an end user through the web chatbot, as reported by GamerGains.
• Before the trusted users of our tool would also receive auto-replies similar to non-trusted users when sending direct messages on Discord to the AwesomeQA bot to contact us. Now they only receive direct replies from our team to their queries to avoid confusion and ensure a smooth experience.
• There was an issue reported by Flash Trade which we fixed where the bot failed to generate a response to a question due to a policy violation error. The error occurred because a whitelisted link in the answer was not recognized due to a case mismatch in the knowledge base. For reference: the bot is configured to never share any links in its replies unless they’re a part of the project’s knowledge base.
• Fixed an issue where the channel category permissions on the bot settings page were not saved correctly when toggled too quickly multiple times during the initial setup (onboarding).
• Resolved an issue where the bot was adding community members' unanswered questions to the knowledge base, leading to unnecessary knowledge items being listed on the knowledge base page of the dashboard.

Top-Tier Security For Your Data

We understand the importance of security when it comes to handling sensitive data. That's why we've partnered with Drata to obtain our SOC2 and GDPR certifications, ensuring that we meet the highest security standards.

These certifications provide peace of mind, knowing that both you and your users can trust our platform to protect your valuable information. With our commitment to security, you can use our tool with confidence, knowing that every measure has been taken to safeguard your data.

Your users will appreciate your dedication to privacy and security when you choose a solution like ours. By prioritizing their safety, you're not just meeting expectations; you're exceeding them.

Other fixes and improvements

• Fixed a bug that caused an interaction error when submitting feedback for an answer using the default buttons from the menu in Discord.
• Resolved an inconsistency between the total and displayed knowledge items in communities where learning from non-trusted users is disabled and ensured that only messages from the question author or trusted users are used in prompts by removing context messages from non-trusted users.
• Fixed an issue where opening a ticket incorrectly updated the ticket preview list with tickets that didn't match the selected filters.
• We now have an additional OpenAI instance as a backup, ensuring uninterrupted service and reliability even if the primary instance goes down.
• There were some web pages scraping issues reported by a user that have now been resolved.
• Implemented additional improvements to the prompt to ensure greater accuracy in the bot's responses.
• Improved the system’s algorithm to enable faster loading of knowledge items in the knowledge hub.
• Addressed and fixed the problem of duplicate knowledge item entries in the knowledge hub that originated from the announcement channel.

Introducing Live Chat & Ticketing For Websites

We are excited to announce the launch of our human live chat feature! Previously, our web chatbot could only provide automated answers to user questions using your knowledge sources. Now, users have the option to speak with a human agent by clicking "Contact Support". This action prompts the creation of a ticket, which agents on the AwesomeQA dashboard can claim.

Agents can reply directly to user queries via the dashboard, with responses sent to the user's provided email if they’re no longer on the website. Users can then return to the chat and ask follow-up questions through a link provided in the email.

This update allows for seamless live conversations with human agents and efficient ticketing through our web chatbot. To enable this feature, simply add our web chatbot to your website using the developer manual and notify us at (hi@pluno.ai) or (Telegram: @syedaqa) to activate the human live chat and ticketing function for you.

Translate Feature for Announcements

Now, every time an announcement is made in your Discord server, the AwesomeQA bot will provide a button for members to translate the announcement into their preferred language. The translation will be a private message, visible only to the member who pressed the button.

This feature will help ensure all members can access announcements in their language while eliminating the need to manually publish announcements in multiple languages in international channels. To enable it, visit this page on your dashboard Translate Announcements, select your announcement channels, add the languages you want to support, and save changes. After completing these steps, the feature will be instantly available for all future announcements.

Customize Bot Responses In Web Chatbot

Noticed the bot's response in the web chatbot needs a bit of tweaking? Now, you can directly review these conversations and create FAQ items around the bot's answers. This allows you to ensure the bot responds exactly the way you want in the future. Simply head to the web chat history page and use the three-dot menu to start customizing answers to your liking.

Other fixes and improvements

• The issue with the Telegram bot displaying markdown links in raw format has been resolved.
• Removed the requirement for "Add reactions" and "Use application commands" permissions. Also, disabled the /contact_us feature since it was only used by community members, not managers.
• We have reversed a recent change aimed at preventing responses that repeated information provided by a trusted user. This reversal was necessary due to a significant drop in the response ratio caused by numerous false positives.
• Updated the FAQ insights to use our latest AI model. As a result, they now incorporate responses to future questions effectively.
• Fixed the issue where deleted source items disabled status reappeared after embedding updates.
• Fixed bot responses to display URLs as plain text instead of hyperlinks.
• The bot now ignores repeated questions from the same user, only responding the first time and remaining silent on subsequent identical queries.
• The bot is now more consistent, providing similar responses when asked the same question multiple times.
• The bot now uses markdown to format its responses on Discord.
• The issue of accidental scrolling in Firefox when opening the web chatbot has now been fixed.

Lightning fast replies coming to the web chatbot

Providing a great user support experience is crucial for all digital products. It often determines whether users will choose your product or go to a competitor. To meet users’ needs directly where they come up, companies like DODO Exchange, Ocean Protocol, and Lava Network have integrated our bot on their websites.

We like pushing the limits, which is why we significantly improved the response time of the AI while keeping the same high quality of responses for optimal user experience.

Feel free to try it out with your data on the dashboard.

Combining multiple messages for more context

If a user sends several messages for one question, our system combines them and replies to the last message. This way, the AI uses all the information the user provides to generate an answer.

You can easily follow this by checking which message the bot replied to.

However, if the bot is tagged in each message, it won't combine them, making it easier to test the bot's replies.

Other fixes and improvements

• The bot no longer responds when tagged by another bot to avoid unnecessary replies.
• The web chatbot can now also be included in the <head> section of your website. Thanks to Lava Networks for bringing it to our attention!
• You can now toggle Learn and Reply permissions for your Telegram channels.
• You can now see the names of your Telegram groups on the permissions and analytics page.
• Fixed an issue when scraping knowledge from Google Docs. Thanks to GRVT for flagging it!
• When the bot is tagged and does not know the answer, it replies with a link for the team to add the relevant knowledge. Once saved, now the bot will also update its answer.
• Fixed an issue to prevent the bot from recommending to open a ticket on Telegram if that option only exists on Discord. Thanks to SingularityDAO for the report!
• Fixed a markdown formatting issue on Telegram.
• Telegram admins are now automatically marked as trusted users as soon as they write their first message after the bot has been added.
• You can now directly add the bot to another Discord server or Telegram group from the settings menu on the dashboard.
• Already authenticated users will no longer have to log in again when inviting the bot to a new community.
• Fixed an onboarding issue where our system would continuously push you through the onboarding flow even after completion.
• Added an additional security measure to avoid the bot mentioning other non-trusted users in its response.
• We now filter non-trusted links before they get to the generative AI step to make sure our bot will never post any malicious links.
• Slightly improved the response time of the community bot.
• Fixed the 🖊️ icon when using the Apps → AQA - Edit command on Discord. Thanks to Ar[dot]io for the report!

GPT-4o is here!

We've upgraded to the latest OpenAI language model, GPT-4o, known for its superior performance and 2x faster responses. The switch has already been completed on our end, so no action is needed from you. We'll be conducting ongoing testing to maximize the benefits of this new model. This upgrade is part of our commitment to enhancing our service for our users.

More control in Telegram

Previously, once the bot was added to a Telegram channel, there were limited options for controlling its behavior. Now, we provide similar controls for TG as Discord. For example, if you prefer the bot not to learn from the TG channel and only respond using knowledge from other sources, you can toggle off the 'learn’ permissions. This can be configured on the bot settings page.

Train it on Tweets now

Users have frequently inquired about training the bot using their X (formerly Twitter) channels. While direct scraping of X profiles is restricted, you can now train the bot with your tweets. Simply share the tweet in the Discord server, and we'll utilize the embedded text to enhance its learning.

No wrong removal of bot messages

To avoid incorrect removal of bot messages, we've adjusted the rating system. Now, bot responses won't be deleted based solely on downvotes if the ratings are from random community members. Deletion occurs only if trusted users or the author participate in the rating process.

Other fixes and improvements

• Fixed a bug that prevented users from removing custom footer text after it was set.
• Fixed an issue where editing outdated information from announcement channels didn't display correctly in the knowledge base. Now, AI generates Q&A formulations for each announcement entry, matching the FAQ format.
• Resolved a bug where the bot would occasionally respond twice consecutively to the same question in chats.

More Ways to Improve Performance

Previously, users could downvote a bot's response to indicate dissatisfaction. However, we lacked clarity on the reasons behind the downvotes, requiring us to speculate based on reviews of downvoted responses.

Now, with the new feedback mechanism, trusted users can select predefined reasons for downvoting and provide specific feedback, enabling us to promptly address issues and enhance the bot's performance. This streamlined process ensures that feedback is received promptly and reduces the risk of it being overlooked.

Next time you downvote a response, you'll automatically see these options available.

Disable DMs server-wide with a click

Scammers often target Discord servers, making members vulnerable to DMs impersonating team members. When attacks occur, server support is flooded with scam complaints. Now, you can disable server-wide DMs with a single click in bot settings. Grant the AwesomeQA bot 'Manage Server' permission to use this feature.

Compatible with non-English servers now

Many servers used languages other than English as their primary mode of communication in channels. When installing the AwesomeQA bot for support, they were only able to train it by uploading web pages like documents. Consequently, the 'chat history' wasn't accessible for the bot to learn from.

Now, the bot has the capability to learn from chat history for all servers, irrespective of whether the messages are in English or not. This feature has been enabled for all servers, providing benefits to every user.

Other fixes and improvements

• Resolved a scraping issue with Google Docs, specifically when redirects were involved.
• If a user downvotes the answer they received, a cooldown period is triggered to reduce further attempts, as subsequent answers are unlikely to improve accuracy.
• Now, you can see the last update time for the web pages uploaded. This helps you know if those pages need to be scraped again due to recent changes.
• Before, users had to delete multiple entries one by one from the knowledge base when different FAQ formulations existed. Now, only one entry appears, making the knowledge item deletion faster.
• The bot now better understands whether it's in Discord or Telegram when answering questions, ensuring context-appropriate responses.
• You can now configure the bot to learn from the FAQ channel in the server, similar to how you could for the announcement channel before.
• Previously, even when tagged, the bot sometimes didn't generate an answer if it deemed there wasn't enough information. Now, it replies with whatever info it has, improving user experience.
• Now, when missing information is added to the knowledge base, the bot's message about the absence of information automatically transforms into an answer to the question.

Enhanced Bot Permissions: Your Control, Your Way

Previously, users faced inconvenience when manually adjusting permissions for each ticket channel created in the tickets category. Now, default permissions can be set for each channel category, simplifying the process.

Plus, users managing multiple servers can now access all permissions and configurations from one account, eliminating the need for multiple logins.

Also, sometimes there were missing channels on the permissions page, as they lacked 'view channel' permissions from the Discord side. With this update, such channels are now visible, enabling users to configure permissions easily.

To top it off, we have also improved the overall loading speed of the page. Visit your permissions page to explore these enhancements.

Compatibility with Additional Document Types

Now you can upload even Google Drive-hosted docs, spreadsheets, and presentations for bot training, which was not possible before. You can try it out in your web pages section right now.

Other fixes and improvements

• The newly added FAQ items are instantly listed and available for all actions now.
• A bug that initially hindered the web page scraping process has now been resolved.
• Fixed the issue where the bot sometimes replied to questions in other languages in English.
• The bot now properly hyperlinked channel names in its responses, allowing users to navigate directly to the channel with a single click.
• Timeout errors reported during the extraction of FAQ insights from communities have been fixed.
• The error notification on the dashboard during the AQA-Edit flow, when adapting answers on the web app and saving, has been fixed.
• Newlines are now correctly formatted during generation. Previously, they appeared as "\n" while generating the answer but were formatted correctly once the generation was complete.
• Implemented a performance improvement to address slow reload times for FAQ items on the Knowledge Hub after editing.
• The issue where the correct dashboard wasn't opened by default through the 'add info' prompt from the channel chat, especially when users had access to multiple dashboards for different communities, has been resolved.
• The bot now notifies you via Discord DMs once the training process is complete, eliminating the need for manual checking.

Introducing the Knowledge Hub

The new Knowledge Base shows the knowledge items from all source channels, i.e. your chat history, any documentation, and your manually added FAQs.

This change also allows you to delete specific knowledge items.

Ever saw the bot using an outdated piece of information? Search for it and delete it with just two clicks.

This feature was developed based on feedback from our customers. Shoutouts to Mickey from The Graph, Ruben from Fuel Network, Shaun from Chainflip, Andy from Consensys, Berm from Verasity, Agustin from Paraswap, Robots Farm, and Sei!

34% faster replies

Think AwesomeQA was already quick? It was faster than most humans. With this update, it's lightning-fast!

We revamped our data structure for storing knowledge, turbocharging the retrieval process.

Result? Response time slashed by an average of 34%.

Other fixes and improvements

• You can now use PDFs hosted by Google Drive using this link format: https://drive.google.com/uc?export=download&id=your-file-id
• Add error message when trying to add a link to Google Docs as documentation source. Soon, we will add support for Google Docs.
• The bot is now able to have longer conversations with users in community channels.
• ‘Docs’ have been renamed to ‘Web Pages’ to be more inclusive of all potential web-based knowledge source: website, blog, documentation, Notion sites & many more.
• Fixed a bug that prevented the bot from responding without the read message permission on Discord.

Instant Feedback for AwesomeQA Users

We understand how important it is for clear instructions to be available as you set up and begin to use the AwesomeQA bot. Now, if AwesomeQA is tagged but can’t reply, you will receive more specific guidance instantly in response to your question:

1. When no information exists: Instructions on how to add to the knowledge base will be given.
2. When ‘reply’ permissions are toggled off: The user asking the question in that channel will be notified.
3. During onboarding: Instructions on how to start the onboarding process will be provided.
4. During training: Information to wait for the training process to complete will be indicated.

Update to Feedback Emoji

We have further enhanced our feedback system by updating our emojis. To offer more clarity, feedback submissions from our trusted users will now include a "Mod certified" text in addition to the emoji. Thanks to ninefungerbils from Consensys for asking for it.

Other fixes and improvements

• We experienced an issue with Telegram that affected the bot's ability to detect messages and learn from trusted users, resulting in some cases where learning was incomplete. This issue has now been resolved.
• We have enhanced the onboarding process with clearer step descriptions, the integration of URL scraping rules into a tooltip on the docs page, the instant opening of tooltips on mobile, and an additional "Finish" button on the summary page.
• Fixed an error happening when clicking on AQA-Add for any message that was not from AwesomeQA, as well as a formatting issue when links were part of the preview.
• We have reintroduced pre-filled answers when you execute AQA-Edit. Earlier, due to a bug, the answers were not being pre-filled after executing AQA-Edit.
• GitHub wikis can now be used as a documentation source.
• Fixed an encoding issue related to the ‘custom footer text’ feature.
• Fixed a bug that sometimes, due to caching, newly added questions from the knowledge base were not answered.
• Bug fixes and performance improvements for discord feedback emojis.

Feedback Buttons on Discord

We changed the feedback mechanism from reactions to buttons so that our bot does not have to vote itself. Further, we believe that the buttons incentivize users to provide feedback and are separated from additional reactions such as '💯', which do not contribute to the rating.

As a result of this change, you are no longer able to view who upvoted or downvoted an answer by hovering over the buttons.

To address this issue, we have implemented a new system based on feedback from cryptopia and 184eth of ensdomains:

1. When a Trusted user, such as an admin, upvotes an answer, a '✅' will appear to confirm the upvote.
2. Similarly, when an answer is upvoted by the author i.e. the person who asked the question, a '👍' emoji will be displayed.

Custom Footer Text

Now you have the option to include a personalized message, such as a disclaimer, in the footer of the bot's responses from the settings page. This message can be up to 200 characters in length. This feature was created based on direct feedback from our customers, including Robots Farm.

AI update

We deployed a new version of our AI which reduces the number of wrong responses. We started focusing more on recent information, and now use all the available context to make a good judgment of what’s helpful and what isn’t.

An Additional Way to Edit Bot Answers

Previously, when you wanted to edit a bot's answer, a link was provided through a hidden message after using the AQA-Edit command. However, there was a Discord bug that prevented some users from seeing this message with the link.

We have now added a clickable '🖊️' edit emoji directly on the message to solve this issue. This means that even if you cannot see the hidden message with the link, you can still easily edit the bot's answer by simply clicking on the '🖊️' emoji.

Other fixes and improvements

• The support portal now shows tickets for all channels with learn or send permission
• Fixed a timeout error that caused to bot ignore a small percentage of messages
• Fixed a bug that caused our bot to not respond in the language of the question
• Toggle all now shows the correct state if all permissions in the category are turned off
• Fixed a bug where our website scraper would only find 1 page
• Fixed a bug that if someone replied to a message more than 12 hours ago, the referenced message was not included as context for the AI
• Fixed a bug that displayed the discord message in the support portal without content