September 14th, 2026
Hi everyone,
You’re likely familiar with this situation: your teams process dozens of requests every week, your AI agents perform exceptionally well, yet when it’s time to take stock, no one really knows where things stand. The data exists, but it’s scattered across chat histories, files, and spreadsheets that no one updates anymore. So you end up manually piecing numbers together or relying on gut feeling.
That’s exactly what we wanted to solve this month, starting in the Knowledge section, specifically within your Knowledge tables. These are your business data points organized in rows and columns—leads, customer inquiries, applications, marketing campaigns—structured so your agents can read and use them to assist you. Until now, you had to fill them in manually.
Not anymore. Your agents now write to them automatically as they work. For instance, an agent qualifying leads will update your sales pipeline while you focus on other priorities. Views adjust to what you want to see, and if you need a specific dashboard, all you have to do is ask.
Your agents populate your tables on their own
An agent can now write directly into a knowledge table. Instead of just reading your data, it enriches it continuously. Every processed request, analyzed document, or conversation automatically becomes a clean row in the right columns.
This addresses a widespread issue: data generated during work often gets lost. Your agent would complete a task brilliantly, only for the details to sink into chat history. Counting, comparing, or identifying trends became nearly impossible. When making decisions, team members either guessed or spent days rebuilding numbers manually.
This workflow applies to any continuous stream of incoming tasks—whether support tickets, marketing campaigns, or job applications. Take an HR department hiring for five open roles simultaneously. As the agent screens incoming resumes, it automatically populates a table with target roles, years of experience, key skills, and availability. No manual entry required. Three weeks later, instead of opening 120 resumes to make a shortlist, the hiring manager reviews an organized database and notices that top applicants are concentrated in only two roles. The other three listings aren't weak, they're under-promoted, an insight that passive reading would never have revealed.
You only see what matters to you
Your tables now support filters and groupings. Select your criteria, and the display reorganizes instantly. You can group rows by column—by owner, status, priority, and more—to digest information in structured sections rather than endless lists.
This grouping functionality also unlocks Kanban views. Each group becomes a column, each row a card, and progress can be reviewed at a glance rather than decoded from a hidden status column. The same table can be viewed in two ways depending on your goal: a grid when you need to compare, count, and analyze; a Kanban board when you want to monitor workflow and identify bottlenecks.
This eliminates scrolling through 800 rows to find the 12 relevant to you. More importantly, it removes the habit of exporting data to a spreadsheet “to sort in peace”, which inevitably created outdated duplicate files.
Consider a customer support manager tracking 800 open tickets from the current quarter. Instead of reviewing everything, they filter by unresolved tickets and group them by issue type. Within seconds, a trend emerges: one-third of complaints stem from a single product line. Volume isn't the core issue—the product feature is, requiring engineering intervention rather than customer support fixes. The same approach benefits executive management: a board filtering by the current quarter and grouping by business unit enters meetings with a shared single source of truth, rather than five disjointed reports.
Dynamic views generate your ideal interface
Directly from the Knowledge section chat, you can request the exact view you need. Describe what you want to track, and Swiftask generates a tailored interface: a dashboard, a stage-by-stage pipeline, or an executive summary view. You don't even need to configure an agent beforehand—a simple sentence in plain language is enough.
This removes friction and turnaround time. Previously, custom reporting meant submitting a ticket to the data team, waiting for delivery, and often receiving a report for a need that had already evolved. Now, you can test a visualization approach in minutes and iterate immediately if needed.
A marketing director allocating year-end budgets can simply request a view crossing acquisition cost, conversions, and ROI by channel over the last six months. She receives a clean dashboard, identifies that two channels consume most of the budget while generating a third of results, and reallocates accordingly. The same mechanism helps track organizational AI usage: a view crossing session counts, error rates, and token consumption by agent highlights which assistants are costly relative to performance, or failing due to vague instructions. You know what to optimize first and have the data to support it.
Your agents confirm details before acting
When a request could be interpreted in multiple ways, your agent no longer guesses. It pauses, prompts you through a clear interface, and offers defined options. Once you select an option, it proceeds with the correct instruction.
The immediate benefit is clear for sensitive operations : sending external communications, modifying database entries, or initiating spending. However, the secondary benefit is equally valuable: team members don't need to be expert prompt engineers to get precise results. The provided options serve as clear guardrails, helping new users learn as they go.
For example, an HR agent tasked with following up on pending candidates will ask whether to target candidates waiting over 15 days or over a month, and whether to include all roles or only open ones. A 10-second confirmation prevents misdirected emails and hours of manual corrections.
Together, these four capabilities create a unified workflow: agents generate your data, filters isolate what matters, dynamic views inform your decisions, and agents confirm intent before taking action.
To explore these features, head to Knowledge ➔ Knowledge Tables in your workspace.
September 6th, 2026
👋 Hello, and welcome back!
Vacations are behind us, schedules are filling up, and your website is about to see traffic pick back up. Perfect timing: the chat bubble you embed on your web pages has just taken a massive leap forward.
It is now built on the exact same foundation as the agent deployed within the main application. Anything you can do inside Swiftask, your visitors can now do directly in the chat bubble. Same features, same possibilities, same smooth experience. No more light version of the widget.
What does this actually mean for you?
This bubble is your Swiftask agent embedded directly on any webpage. Anyone who clicks on it needs no account, no setup, and no downloads: they open it, ask their question, and get an answer. Here are a few typical use cases:
A Support Agent on your website: A customer runs into an issue at 10 PM on a Sunday. Instead of waiting until Monday morning, they open the bubble, describe their problem, attach a screenshot, and walk away with a solution. Your support team only handles what truly requires human attention.
An HR Agent on your intranet: "How many days of paid leave do I have left?", "How do I report sick leave?", "Where can I find the remote work request form?". These questions pop up endlessly and pull your HR team away to handle low-value tasks. An agent connected to your internal documentation answers instantly, 24/7, without anyone having to open a support ticket.
A Public Assistance Agent: For local governments, public administrations, or non-profits: the bubble answers citizens' questions regarding administrative procedures, opening hours, or required documents. Accessible to everyone, no account required, available around the clock.
A Pre-sales Agent on your product pages: A prospective client hesitates and has three quick questions they would never bother submitting through a contact form. The bubble captures that exact moment. They get their answers, and you capture a lead that would have otherwise left without a trace.
Real-time streaming responses
This used to be a major pain point. Until now, visitors would send a message… and stare at a frozen screen. Five, eight, ten seconds of total silence, and then the entire response would drop at once.
The issue? That visitor doesn't know you yet. Faced with a frozen screen, they don't think "It's processing"; they think "It's broken". So they either resend the message or close the tab altogether.
Now, the response streams in right before their eyes, word for word, from the very first second. They immediately see that it's working, start reading right away, and waiting becomes invisible.
See the Agent at work
Your agents don't just generate text answers: they search the web, consult their knowledge base, query connected tools, or send emails. These actions take time, and until now, they happened entirely in the dark.
Now, your visitors can follow the work in progress in real time. Whenever the agent triggers a skill, the action displays step-by-step as it happens.
This turns waiting into an engaging experience: waiting ten seconds while seeing "Searching internal documents..." feels entirely different from waiting ten seconds in front of a blank screen. It also builds trust: an employee who sees the HR agent consult the official company policy before answering grants that response far more credibility. It's no longer a black box making assertions; it's an assistant demonstrating its work.
Well-organized conversations
Previously, everything was stacked into a single, continuous thread. A returning visitor with an unrelated question saw their new request tacked onto the old discussion. The agent carried all the previous context along, running the risk of giving an off-topic answer.
The widget now supports full session management, just like the Swiftask app. Each conversation lives in its own dedicated space, complete with its own history and context.
Imagine a customer who wrote to you on Monday regarding an order and comes back on Thursday with a billing question: two distinct threads, two clean contexts, and they can easily jump back to Monday's exchange if needed.
Voice conversations
Your visitors can now speak directly to your agent instead of typing. With a single click on the microphone icon, they ask their question out loud, and the agent understands and responds.
While it might seem like a minor addition, it is actually a game-changer. The vast majority of your visitors arrive from mobile devices, where typing long, detailed messages can be tedious. Using voice, they can explain their situation in fifteen seconds with all the necessary details—where they might have otherwise typed just three keywords. Your agent receives a complete request and provides a much better answer.
This is critical for public service agents or field operations: a user who struggles with written text, a technician with their hands full, or a senior citizen navigating an administrative form. Voice opens the door to everyone left behind by the keyboard.
File sharing
Your visitors can now attach files directly from their computer or mobile device into the conversation. No need to switch to email or submit a separate form: everything is handled within the same chat, and your agent works directly on the received file.
A screenshot of an error message for your support agent. A pay stub to analyze for your HR agent. A photo of an administrative letter for your public assistance agent. A project specification document for your pre-sales agent. In every case, the conversation kicks off with the actual document right in front of the agent rather than a vague description.
A bubble that matches your brand
The selection of icons for the floating action button has expanded significantly: you now have a wide array of icons to choose from to fit your brand identity. It’s a subtle detail, but it’s also the very first element your visitors see of your agent—it might as well look like you.
Where to set this up?
Navigate to Agents ➔ Select an Agent ➔ Configure Agent ➔ Chat Bubble (Widget) ➔ Integration or Appearance.
Already have a widget deployed live on your site? Everything updates automatically! Still, take a quick trip over to your website and test asking your agent a question—experiencing the new flow from your visitors' perspective is the best way to appreciate the upgrade.
September 2nd, 2026
Back-to-school on Swiftask! We've prepared some great new updates for you.
A new season means new challenges and projects picking back up at full speed. To help you navigate this September with peace of mind, we’ve built some exciting improvements into your workspace: enhanced security, easier-to-customize assistants, and new AI models to explore.
Here is a practical overview focused on what this changes for you day-to-day:
1. Access History for Shared Resources
The sharing settings now feature a dedicated audit tab accessible everywhere across Swiftask—whether for an agent, a resource (document, folder), an artifact, and more. It automatically logs every access attempt and permission change.
You maintain full visibility over the governance of your data and tools. At any time, regardless of the type of shared item, you can track permission changes (such as a role upgrade from Viewer to Editor, the exact timestamp, and who made the change) to keep your environment secure.
Whenever you open the sharing options for an agent, document, or artifact, head over to the Access History tab. You'll find a complete chronological list of role modifications made by your team, ensuring every access right remains under control.
2. Simplified Custom Agent Duplication
A new duplication interface allows you to clone an existing agent while selecting exactly which elements to keep.
Save precious time when creating new assistants. No need to reconfigure everything manually—start from a proven baseline and adapt it in seconds.
Use Case: To adapt a "France Customer Support" agent for the Spanish market, simply click Duplicate, preview its configuration, and swap out only its knowledge base with your translated documents. Your new assistant is ready to use immediately.
3. Centralized Agent Monitoring (Session-Centric Monitoring)
A unified dashboard now brings together all open chats with an agent alongside its error logs, helping you continuously optimize assistant performance. This bird's-eye view makes it easy to understand how your agents are being queried and instantly spot potential roadblocks in their responses.
For example, from an HR agent's monitoring tab, you can filter conversations to review a session that triggered an error, inspect the details drawer, and tweak your instructions to perfect future responses.
4. Microphone Selection During Meetings
You can now view and choose the specific audio device used during your Swiftask meetings or recordings.
Avoid audio mishaps and recording interruptions. If you use multiple devices (Bluetooth headset, wireless earbuds, or built-in computer mic), you stay in control of your audio source and can fix mic issues on the fly.
When starting a meeting (Meetings ➔ Start a Meeting), your device name (e.g., AirBuds 8) appears directly on the control bar. Click Settings to switch to another connected mic without having to restart the recording or close the session.
5. Seamless Reconnection & External App Verification (Outlook & Microsoft 365)
Swiftask now proactively checks the status of your connections to external apps (such as Outlook by Swiftask or Microsoft 365 by Swiftask). If your session or access token has expired, the interface detects it automatically.
This prevents unexpected interruptions and frustrating error messages midway through a chat with your assistants. Instead of an agent getting stuck claiming it no longer has access to your emails, you are notified instantly to re-authorize access.
If your Microsoft account connection expires or is revoked, a clear banner appears directly in your conversation thread. Simply click to renew your connection in seconds and resume your work without restarting the session.
6. Enhanced IdP (Identity Provider) User Provisioning
Automated synchronization between Swiftask and your enterprise identity providers (such as Okta, Entra ID / Azure AD, or Google Workspace) is now smoother and more precise.
IT teams benefit from simplified, secure administrative management. Assigning and revoking employee access and roles happens automatically during onboardings and offboardings, with zero manual input required.
When a new employee joins the team, their Swiftask account is automatically provisioned with the correct permissions as soon as they log in via your company portal.
7. Arrival of 3 New LLM Models
Our model catalog expands with the addition of Claude Sonnet 5 (via OpenRouter), Scaleway DeepSeek V4 🇫🇷🇪🇺, and Gemini 3.7 Flash.
You now have the right tool for every specific need—whether you prioritize speed, advanced programming capabilities, or strict data sovereignty.
If you are working on summaries of confidential documents, choose Scaleway DeepSeek V4 for fast processing hosted on European/French clouds. If you are developing complex scripts, assign Claude Sonnet 5 to your dedicated coding agent.
All these features are already available in your workspace! If you have any questions or need guidance getting started, our entire team is here to help.
August 20th, 2026
Hello,
To give you even finer control over your projects and automations, we've just added 6 brand-new, cutting-edge language models to our catalog.
Whether you're chatting directly or leveling up your custom AI Agents, these new LLMs bring faster speed, massive memory windows, and outstanding reasoning capabilities.
Here is a quick guide to help you pick the right “brain” for your needs:
➡️ Claude Sonnet 5 (Google Cloud Platform) 🇪🇺: The coding and autonomous workflow specialist
If you need to audit a complex codebase, fix bugs, or build automation scripts, this model automatically adjusts its reasoning level to carry out multi-step technical projects smoothly. It’s also a reassuring choice if you handle sensitive work data, as its deployment via Google Cloud EU infrastructure guarantees fully secure hosting and processing within Europe.
➡️ Scaleway DeepSeek V4 Flash 🇫🇷🇪🇺: Lightning speed on a sovereign cloud
If you regularly deal with lengthy legal documents, project specifications, or hundreds-of-pages-long reports, this model leverages its massive memory window (1 million tokens) to generate summaries or retrieve specific information in record time. Since it is hosted directly in French data centers on the Scaleway sovereign cloud, you can process your most confidential content with complete peace of mind.
➡️ Qwen 3.8 Max: Massive memory for long-term workflows
With an incredible processing capacity (up to 1 million tokens, equivalent to thousands of pages), this model is your ultimate partner for analyzing huge document archives or managing developer agents working over several days.
➡️ Mistral Large 3 & Medium 3.5: European excellence and visual analysis
Mistral Large 3: Fluent in over 40 languages, it effortlessly analyzes your largest financial, legal, or technical reports in a flash.
Mistral Medium 3.5: Versatile and agile, it handles both text and visuals (charts, diagrams, screenshots) while letting you adjust its reasoning depth depending on whether you need a quick answer or a deep dive.
➡️ xAI Grok 4.6: The specialist for long working sessions
Designed to never lose track, Grok 4.6 excels at managing lengthy projects and creating visual interfaces. It’s ideal for building prototypes, refactoring code, or automating complex step-by-step financial analyses.
How to start using them today?
To chat directly: Go to the Chat and select your new model from the “Change AI” drop-down menu.
To set up your assistants: Go to the Agents section ➔ Configure Agent ➔ Model Settings (LLM) to assign these new capabilities to your custom agents.
August 14th, 2026
Hello!
We're continuing to improve Swiftask every day, and here’s a look at the great new features we’ve prepared for you this week.
1. Synchronization between the Knowledge and Projects modules
When you managed your task notes inside your projects, they used to remain isolated as simple working notes.
The good news is that your projects now interconnect in real time with the Knowledge module.
➡️ Your task notes become real sources of knowledge, ready to be used directly to power your Chat conversations or train your AI agents.
➡️ Any creation or modification made in your projects is automatically synced with the Knowledge module, ensuring your entire team always has up-to-date information.
Imagine creating a new project for a product launch and adding key task notes to it. Automatically, these notes appear in the Knowledge module. The entire team—including new members added to the project—can directly query the AI using this information or use it to build a dedicated agent, all while maintaining secure access permissions (new members can view the data without being able to modify it).
How to try it?
Go to the Projects section, create or edit a project with task notes, and open the Knowledge module to see your data instantly synced and ready to use!
2. Governance settings for primary and fallback LLM models
When configuring your AI agents, you could define a primary language model (LLM) as well as a fallback model, but managing the latter sometimes lacked flexibility.
Now, you have complete control over your fallback models thanks to new governance options.
➡️ Take full control: You can now choose to enable or remove a fallback LLM based on your specific performance or budget requirements, using the new "Remove fallback LLM" button.
How to try it?
Go to your agent settings (via the Agents tab), navigate to the model parameters (LLM), and use the new "Remove fallback LLM" button.
3. Direct connection to the entire Microsoft 365 suite
Access to your Microsoft tools was previously a bit limited. That is now a thing of the past: you can directly connect your Microsoft 365 account to Swiftask!
➡️ Your whole workspace in one place: Your agents securely access the apps you use every single day (Outlook, Calendar, Teams, SharePoint, OneDrive...).
➡️ One-click document creation: Your assistants no longer just read your data. They can directly create real Word files (.docx), Excel spreadsheets (.xlsx), or PowerPoint presentations (.pptx).
➡️ In practice: Preparing a project review? Your agent gathers the information, generates the PowerPoint presentation, and instantly sends it via email to your team using your Outlook account.
Where to find this feature?
Go to Agents ➔ Edit ➔ Skills, then search for Microsoft.
4. Improved automatic text recognition (OCR) for large-format documents (A3)
Do you often work with blueprints, technical diagrams, or large financial spreadsheets? Great news: we've significantly boosted our text recognition technology (OCR) for these complex formats.
➡️ Much more accurate reading: A3 files and highly detailed visual documents are now processed effortlessly.
➡️ Better Chat responses: No more misinterpretations or glitches when you drop a large document into your conversations.
➡️ Comprehensive Knowledge integration: Once added to your bases, these large files are seamlessly processed and understood by your AI agents.
Want to try it out?
Simply drop an A3 file or a complex document into the Chat or Knowledge section, then ask the AI your question—it will find the exact information for you in no time!
5. Outlook trigger fix and agent trigger history interface improvement
Your automations sometimes ran into glitches when receiving emails, and tracking the details of automated executions wasn't always straightforward. All of that is now resolved!
➡️ 100% reliable email automation: The issue with the Outlook trigger has been fixed. Your automated workflows triggered by incoming emails are back to running seamlessly without interruption.
➡️ Clearer daily tracking: The history interface has been redesigned to give you complete visibility into your assistants' activity. Spot at a glance when an agent was triggered, what it executed, and whether it encountered an error.
➡️ In practice: If you set up an agent to automatically process incoming email requests, you can now monitor every step of its work in real time directly from your dashboard, with full transparency.
Where to find this feature?
Go to the Agents section, select your assistant, and click on the History tab to check out the new interface.
6. Direct task execution on your computer with Computer Access
Until now, the AI remained limited to the Chat: it could explain steps or generate code, but it couldn't act directly on your machine (local files, installed apps, browser, or terminal).
Now, the Swiftask desktop app allows your assistant to perform real actions on your computer!
➡️ No more manual steps: No need to copy-paste scripts between the Chat and your computer, or manually carry out what the agent planned (opening a file, launching an app, or filling out a form).
➡️ No more constant recontextualization: Forget about copying and pasting file contents or terminal outputs into the chat.
➡️ Less friction, more delegation: Save time by delegating actual tasks (instead of just getting text responses), while keeping total control through a dedicated per-chat toggle and strict permission settings.
➡️ Imagine needing to analyze a local Excel file (ventes.xlsx) in your Downloads folder to generate a Word report on your Desktop (rapport-ventes.docx). Without this feature, you'd have to open the file, paste excerpts into the Chat, run commands yourself, and copy the result. With Computer Access, the agent acts directly on your machine in just a few seconds!
7. Group synchronization and automatic SCIM 2.0 user provisioning
Managing team member access and organizing your workspace groups takes a major step forward.
Thanks to the integration of the SCIM 2.0 standard, your entire user directory now automates end-to-end with your identity provider (IdP).
➡️ Real-time automatic provisioning (SCIM 2.0): Connect Swiftask directly to your usual identity provider (Entra ID, Okta, Google Workspace, or generic SAML). Account creations, updates, and deactivations sync instantly without lifting a finger.
➡️ Automated offboarding: As soon as a team member leaves your organization or is removed from your IdP, their access is immediately revoked and their active sessions are closed. Keep your workspace secure while instantly freeing up licenses.
➡️ Seamless group and role mapping: No more obscure technical IDs like the old "Azure Group ID"! Your groups now display under their actual names, and you can map an external group directly to a Swiftask role (Admin, Manager, Creator, Member) to adjust permissions automatically.
➡️ Simplified team onboarding: Upon their very first SSO login, new hires automatically join the workspace, inherit the right groups, permissions, and credit limits. They get instant access to the agents shared with their team.
➡️ Multi-domain support & forced SSO: Register multiple email domains for your organization and enforce SSO login to ensure maximum security.
Where to find this feature?
Go to Admin ➔ Workspace ➔ Provisioning section to follow the step-by-step guide, map your groups, and view the audit log.
August 10th, 2026
Hello!
When creating an agent using AI assistance, you were previously guided by a standard default assistant.
The good news is that you now have a choice! You can replace that default assistant with one of your own custom agents.
➡️ No more wasting time explaining everything from scratch. You can now entrust the creation of your future assistants to an agent that already knows your business, brand guidelines, and workflows inside out.
➡️ For example, imagine you have an “Editorial Guidelines Expert” agent trained on your tone of voice, communication rules, and vocabulary.
If you want to build a brand-new agent to write product descriptions:
Open the Agent Builder and select your “Editorial Guidelines Expert” as your guide.
It will directly apply your brand style throughout the entire setup process.
No need to re-enter your guidelines—it leverages its existing knowledge to configure the new agent.
How to try it?
Go to Workspace Settings ➔ Default Agent and AI to select your preferred agent.
Then, simply head to the Agents tab and open the Agent Builder to see the change in action and start building your next assistants.
August 9th, 2026
Hello,
We have just integrated the four latest AI models on the market. You can now choose the exact intelligence that powers each of your agents.
1. Claude Opus 5 (Anthropic)
If you are tired of double-checking an AI's work to fix minor details, you are going to love Claude Opus 5. Its standout feature is its agentic persistence: when handling long or complex tasks, it spots its own oversights and self-corrects before delivering the final result.
It is the ideal partner for designing a strategic plan from A to Z, refactoring code, or drafting in-depth documents. It also features a Fast Mode that executes your requests 2.5 times faster, without unnecessarily flagging your professional content.
2. Gemini Flash 3.6 (Google)
Here, the goal is to move fast while staying ultra-efficient. Gemini Flash 3.6 was built for those who need instant answers, without any fluff or wasted resources.
It consumes significantly fewer tokens and gets straight to the point in as few steps as possible. Pair it with a customer support agent, a help desk chat, or a bot designed to process dozens of emails in a row: answers arrive with zero delay.
3. Kimi 3 (Moonshot AI)
We have all hit a model's memory limit when dealing with a massive document. With Kimi 3, that constraint is a thing of the past thanks to its ability to process up to 1 million tokens at once.
You can feed it an annual report hundreds of pages long, a project's full chat history, or an entire codebase without ever overloading it. A very handy bonus: it can also directly generate interactive assets, such as presentations or comprehensive visual reports.
4. Grok 4.5 (SpaceXAI)
Trained by observing the workflows of real developers, Grok 4.5 boasts remarkable logic for anything involving structure, data, and programming.
It sequences actions very naturally and allows you to adjust its reasoning effort (low, medium, or high) based on the problem's complexity. It is the dream engine for automating repetitive administrative tasks, writing scripts, or cleaning up your data files.
How to update your agents
The setup takes less than 30 seconds:
Go to your Swiftask workspace and open your Agent settings.
In the model (LLM) dropdown menu, select the new AI of your choice.
Save: your agent instantly adopts its new capabilities!
August 5th, 2026
Hello!
It's update time ! We are thrilled to share the latest improvements deployed on Swiftask this August 4, 2026. Our goal is to make your everyday work even simpler and more comfortable.
Here is a breakdown of what this update changes for you:
✍️ A modern text editor for your Branding Kit
What’s changing: Say goodbye to plain, boring text! You can now format your Branding Kit texts just like a real document: add bold, italics, create bulleted lists, or insert line breaks.
What it's for: The Branding Kit helps you define your company's identity (your tone of voice, key messages, words to use or avoid) so that your AI agents always speak with your brand's voice. With this new editor, writing and structuring these guidelines is a breeze, keeping them clear and easy to read for your teams.
What it brings to your daily life: Pure writing comfort. Your editorial guidelines are clean and well-structured, and your agents continue to follow them to perfection.
How to get started: Simply head over to your Workspace Settings → Branding Kit to start formatting your texts.
📊 Clearer and more readable Knowledge Bases (Knowledge Tables & Sources)
What’s changing: Knowledge Tables (those structured data tables, very similar to CSV files embedded in Swiftask) and the detail pages of your information sources (files, websites, SharePoint...) are getting a major facelift with a completely redesigned layout.
What it's for: These tools allow you to feed your agents with precise data (product catalogs, contact lists, FAQs...). Thanks to this visual redesign, navigating through this data and checking at a glance which information is powering your agents' responses is much easier.
What it brings to your daily life: Less eye strain and total control over your agents' "memory". You instantly understand where the information comes from, and you can manage your data tables with much more fluidity.
💻 Your desktop app updates all by itself!
What’s changing: The Swiftask desktop application now supports automatic updates.
What it's for: No more downloading new versions manually from the website or wondering if you're running the latest build. Swiftask takes care of everything in the background.
What it brings to your daily life: The peace of mind of always enjoying the very latest features, security patches, and stability improvements—without lifting a finger.
How to get started: You don't have to do a thing! Just let the app do its magic next time you launch it.
🛠️ Stronger connections and smoother exchanges (Outlook, files, and chat)
What’s changing: Our team has strengthened several elements behind the scenes:
The Outlook connector is now much more robust to prevent unexpected disconnections.
File processing by sub-agents (the specialized components working in the background to handle complex tasks) is now much more reliable.
The chat space with agents has been optimized to be more responsive.
What it's for: These technical fixes are designed to make using Swiftask completely smooth and seamless.
What it brings to your daily life:
Peace of mind: Your automations linked to your Outlook emails won't stop working due to an expired session. No more repetitive reconnection requests!
Efficiency: When you send documents to your agents, analysis is smoother, with a much lower risk of freezes or failures.
Speed: Your chat exchanges are more responsive and stable, even during long discussions.
💡 In short, what this update changes for you:
More comfort: Simplified brand guidelines editing and much more readable data tables (similar to CSV files).
More peace of mind: A desktop app that updates itself and an Outlook connection that stays active.
More fluidity: Faster, more reliable agents to analyze your documents and reply to your messages.
July 24th, 2026
Hello,
We would like to follow up with you regarding the slowdown experienced on the Swiftask platform on July 23, 2026, between 2:00 PM and 9:00 PM (UTC), which affected all our users.
What happened
After a thorough investigation by our technical teams, we identified the source of the issue: in certain configurations within the multi-agent module, an agent and its sub-agents could trigger a loop by calling each other repeatedly. These repetitive executions gradually saturated our database connections, which slowed down the entire platform.
We want to reassure you: no data was compromised, lost, or exposed. This incident was strictly performance-related.
What we have done
As soon as the cause was identified, our teams worked to deploy fixes, which were put into production on July 23 at 9:00 PM (UTC). These improvements include:
Implementing an anti-loop safeguard, which automatically detects and interrupts recursive calls between an agent and its sub-agents;
Optimizing our caching system to reduce the load on our database;
Patching our parallel processing management for better load distribution.
What we continue to do
We are actively continuing our work to strengthen the platform's reliability over the long term:
At the application level: reinforcing early detection mechanisms for agent loops;
At the infrastructure level: reinforcing our capacities and enhancing monitoring of database connections to detect such situations before they impact your experience.
We are fully committed to ensuring a stable and high-performance service, and we will keep you updated on the progress of these improvements.
Thank you for your trust and patience.
Sincerely,
The Swiftask Team
July 15th, 2026
Hello,
We wanted to follow up with you regarding the recent incidents and slowdowns you may have experienced on the Swiftask platform.
What happened
After a thorough investigation by our technical teams, we identified that these incidents were caused by infrastructure issues triggered during peak load periods. These peaks are directly tied to the strong increase in platform adoption over the past few weeks, which put pressure on certain components of our architecture: application services, database, and computing resources.
What we've done
As soon as we detected these issues, our teams mobilized to diagnose the root cause and deploy fixes, which went into production last night. These improvements include:
Optimization of the agent engine deployment to prevent interruptions during execution
Increased database resources to better handle read/write load
Expanded server capacity across our infrastructure after coordinating with Scaleway to increase our quotas
What we're continuing to do
We're actively pursuing optimization efforts at multiple levels:
Application level: load management, processing optimization, and implementation of resilience mechanisms
LLM support optimization: improving resilience for Scaleway and other LLM providers
Infrastructure level: auto-scaling, enhanced monitoring, and resource management
Additionally, to ensure long-term platform reliability and performance, we've engaged an external infrastructure specialist to conduct a comprehensive audit of our systems and guide us through implementing improvements across the entire Swiftask platform.
We're committed to delivering a stable and high-performing service, and we'll keep you updated on our progress.
Thank you for your trust and patience.
Best regards,
The Swiftask team