Changelog
We're constantly working on new features and improvements. Here's what's new with Langdock.
GPT-6 Luna, GPT-6 Sol and Opus 5.5
GPT-6 Luna, GPT-6 Sol, and Claude Opus 5.5 are now available on Langdock! 🚀
GPT-6 Luna and Sol bring stronger capabilities for professional work, coding, and computer use. Luna is designed for focused, high-volume tasks, while Sol is the balanced option for complex coding and agentic workflows. Read more in OpenAI’s official announcement.
Claude Opus 5.5 is the first model in the Claude 5.5 family. It delivers stronger capabilities for agentic coding, knowledge work, and communication than Opus 5. Read more in Anthropic’s official announcement.
All three models are hosted in the EU and available to every workspace. They have been automatically enabled where the corresponding previous model was active. Admins can manage model access in workspace settings.
Composable Workflows
You can now build larger automations in our workflows product by connecting multiple workflows. A workflow can now call another workflow, wait for its result, and continue with its output.
We also added more ways to trigger workflows, made workflow runs faster, and improved how you monitor your workflows and recover failed runs.
Calling other workflows
Select another workflow in the editor, map its inputs, and use its result in the steps that follow. This lets you reuse shared logic across automations instead of rebuilding the same steps.
Nested runs are visible in the run history with Called by and Open Child Run indicators. Workflows can be nested up to five levels deep, and cycle detection prevents recursive calls.
Multiple triggers
You can now add multiple triggers to a single workflow, so the same automation can run from different sources. All existing trigger types can be combined without creating and maintaining separate workflows.
Monitor and rerun workflows
The Workflow Monitor now surfaces invalid connections before a run fails and lets you reconnect them directly. You can also approve or reject pending reviews and agent inputs from the monitor.
Published workflow failures and spend limit alerts can notify recipients through their inbox. Failed runs show when a retry has succeeded, and you can rerun up to 20 failed runs at once, either from the point of failure or from the beginning.
Over the last few weeks, we also shipped a set of smaller improvements across workflows, which you can see below.
- Environment variables: Define shared values once and reuse them across workflows, keeping configuration and secrets consistent without hard-coding them in every node.
- Export workflow runs: Export run data to analyze executions outside Langdock or keep an offline record.
- Import from Zapier and Make: Import existing automations from a clearer entry point, with more useful schema details when an import fails.
- Code node file access: Let code nodes read files produced by earlier nodes, including attachments and generated files.
- Transcribe audio node: Turn audio into text as part of an automation without leaving the workflow builder.
- Webhook triggers: Accept GET webhook requests, optionally require authentication, and preserve raw payloads and request headers when you need the unparsed request.
- Test run context: Test runs retain context from nodes that have already run, so iterative testing does not lose earlier results.
- Output values redesign: We now display outputs in a more friendly way, making it easier to work with node outputs.
Plugins
Plugins bring Langdock’s growing collection of integrations, skills, and tools together in one place, making it easier to discover and use the right capabilities in every conversation.
Add a plugin directly from the chat input to give a conversation access to its context and actions. This lets you use integrations, skills, and tools as you work, without leaving the chat to configure them elsewhere.
Open Plugins from the sidebar to browse the integrations connected to your workspace and discover available skills. Installed integrations appear first, while categories and workspace popularity make it easier to find what you need. Each integration also includes prompt starters, so you can quickly understand what it can do and put it to work.
A simpler home for skills
Plugins also give skills a simpler home. Browse skills alongside integrations, then enable the ones relevant to your work.
To make that experience more straightforward, skills now have one clear state: enabled or disabled. When a skill is enabled, it is ready to use, so you no longer need to install it before activating it. This makes it easier to try skills as they become available and keep only the ones your team needs enabled.
- Agent archiving: Archive agents you no longer actively use to keep your main overview focused while preserving their configuration for later. By default, agents inactive for three months are scheduled for archiving with a seven-day grace period.
- Working folders in Agent Builder Chat: You can now set or remove an agent’s working folder directly from the Agent Builder chat, making it easier to configure the agent while you build it.
- Faster agent-editor configuration: Managing attached files, folders, skills, templates, and integrations is now faster, with a consistent remove button on every item, visible refresh and folder-preview actions that let you switch folders in one click.
- Per-agent action confirmations: Action-confirmation controls are available in the Agent Builder, giving you reliable control over which actions require user confirmation before execution.
- More control over agent questionnaires: Sending a message while an agent questionnaire is open now dismisses the questionnaire and sends your message directly.
- Work in a folder for existing chats: You can now scope an existing conversation to a specific folder with Work in a folder, helping the model focus its responses on the relevant content.
- Clearer context-optimization progress: The context-optimization indicator in chat now matches the visual style of other tool calls, with a progress ring while running and a checkmark when complete.
- View used Skill details: Click a Skill used in a conversation to view its details.
- Skill integrations in ZIP exports: Integrations attached to a Skill are now preserved when exporting and importing Skill ZIP archives. Importers can preview and adjust them before creating the Skill.
- Larger Library uploads: Upload up to 100 files at once when adding documents to your Library or knowledge base. This doubles the previous limit.
- Cleaner file outputs: File outputs now show only the final selected files instead of every intermediate working file.
- Scheduled tasks API: Create, list, update, pause, resume, run, and delete scheduled tasks through the public API.
- Persistent scheduled-task templates: Scheduled-task templates remain visible in the task list after you create a task, so you can create additional tasks from a template at any time.
- Mark all scheduled-task runs as read: Clear all unread runs for a scheduled task with a single click.
- More reliable presentation generation: PowerPoint generation now automatically detects and flags layout issues, such as missing content, very small text, and image-only slides, before delivery.
GPT-Image 2.5
OpenAI's GPT-Image 2.5 is now available on Langdock! 🚀
Today, we are adding GPT-Image 2.5 to Langdock. It comes in two versions:
- Flare is the default model, with improvements in image quality, editing, and speed.
- Sunburst is a premium option for visual work that needs greater control and precision.
You can read more about the launch in OpenAI’s official announcement.
The models are hosted in the EU and are now available to all workspaces. We have automatically enabled GPT-Image 2.5 Flare in workspaces where GPT-Image 2 was previously activated.
Workspace admins can manage model access and enable Sunburst in workspace settings.
Langdock Desktop App
The Langdock Desktop App is now available for macOS and Windows. It has its own icon in your dock or taskbar, so you can open Langdock directly from your desktop whenever you need it.
Log in with your Langdock account to access your existing chats, agents, and workflows. No additional setup is required. The app also lets you work on multiple tasks in parallel with multiple tabs and use keyboard shortcuts for common actions.
Download the app
Download the Langdock Desktop App for macOS or Windows here.
GPT-6 Astra
OpenAI’s GPT-6 Astra is now available on Langdock! 🚀
Today, we are adding GPT-6 Astra to Langdock. It is OpenAI’s most capable flagship model yet and a step up from GPT-5.6 Sol for complex reasoning, coding, computer use, browsing, science, and professional work.
Read more about the launch in OpenAI’s official announcement.
The model is hosted in the EU and is now available to all workspaces.
To get you started quickly, we have automatically enabled GPT-6 Astra in workspaces where a previous GPT version was activated. Admins can manage model access in workspace settings.
Gemini 3.8 Flash
Google's Gemini 3.8 Flash is now available on Langdock! 🚀
Gemini 3.8 Flash is Google's most intelligent Flash model yet, built for long-horizon software engineering, autonomous agents, and complex enterprise workflows. It improves on Gemini 3.7 Flash in coding, reasoning, and knowledge work. Learn more in Google's official announcement.
The model is hosted in the EU and is available in all workspaces where a previous Gemini model was enabled. Admins can manage model access in workspace settings.
Claude Fable 5.1
Anthropic’s Claude Fable 5.1 is now available on Langdock! 🚀
Claude Fable 5.1 is Anthropic’s most capable generally available model for coding and knowledge work, built for complex problem-solving, long-running tasks, and advanced research workflows. Learn more in Anthropic’s official announcement.
Fable 5.1 is hosted in the EU and available to all workspaces. It is enabled automatically in workspaces where Fable 5 was already activated.
Admins can manage model access in the workspace settings.
Gemini 3.7 Flash
Google's Gemini 3.7 Flash is now available on Langdock! 🚀
Today, we are adding Gemini 3.7 Flash to Langdock. It is Google's most capable Flash model yet for coding and agentic workflows, with improvements in software engineering, web development, and knowledge work.
You can read more about the launch in Google's official announcement.
The model is hosted in the EU and is now available in all workspaces where a previous Gemini model was enabled. Admins can manage model access in workspace settings.
Langdock for Outlook
Today, we’re introducing Langdock for Outlook, available on the Microsoft Marketplace. Langdock is now integrated directly into your Outlook inbox and calendar.
Why we built this
Many people spend a significant part of their day managing emails and meetings in Outlook. We wanted Langdock to be available where they already communicate, plan their time, and prepare for meetings.
With Langdock for Outlook, users can summarise threads, draft replies, organise their inbox, manage calendar events, and prepare for upcoming meetings. The same models, knowledge bases, integrations, skills, and workflows available in Langdock are accessible within Outlook.
This makes it easier to keep track of emails and meetings, find relevant information, and work with the right context in one place.
What you can do
- Understand email threads: ask questions about long conversations and turn them into clear summaries with key decisions, open questions, and next steps.
- Organise your inbox: identify important emails, group related messages, and prioritise what needs your attention.
- Draft and improve replies: write context-aware responses, adjust the tone, translate messages, or make your writing clearer.
- Manage your calendar: check your schedule, find available times, and create or reschedule events.
- Prepare for meetings: use related emails, calendar details, and company knowledge to get the context you need before a meeting.
- Use Langdock in Outlook: access your company’s knowledge bases, integrations, skills, and workflows directly from your inbox and calendar.
Availability
Workspace admins can enable Langdock for Outlook in the workspace integration settings.
Once enabled, users can install it from the Microsoft Marketplace and start using Langdock directly in Outlook.
Check out our documentation to get started.
Claude Opus 5
Anthropic’s Claude Opus 5 is now available on Langdock! 🚀
Today, we are adding Claude Opus 5 to Langdock. It is a significant step up from Opus 4.8 for complex reasoning, agentic coding, professional knowledge work, and long-running tasks.
You can read more about the launch in Anthropic’s official announcement.
The model is hosted in the EU and is now available to all workspaces.
To get you started quickly, we have automatically enabled Opus 5 in workspaces where a previous Opus version was activated. Admins can manage model access in workspace settings.
GPT-5.6
OpenAI's GPT-5.6 is now available on Langdock! 🚀
Today, we are adding the GPT-5.6 model family to Langdock. It comes in three versions:
- Sol: OpenAI's new flagship, with state-of-the-art results on coding and agentic search
- Terra: a balanced model for everyday work
- Luna: the most cost-efficient option
All three are a significant step up from GPT-5.5 on knowledge work, reasoning, and long-running agentic tasks.
You can read more about the launch in OpenAI's official announcement.
The models are hosted in the EU and are now available to all workspaces.
To get you started quickly, we have automatically enabled all three GPT-5.6 versions in workspaces where GPT-5.5 was previously activated. Admins can manage model access in workspace settings.
Chat Agent Builder
Creating a new agent in Langdock used to mean filling out a form: writing instructions, picking a model, connecting integrations, one field at a time. Now you can just describe what you want to build in plain language, and the agent builder will take care of the rest.
When you open the agent editor, you'll find a new conversational interface alongside the existing form. Tell it what your agent should do: "I want an agent that monitors my inbox and creates Jira issues from customer requests." It will propose a plan, ask any clarifying questions it needs, and then configure the agent for you: name, instructions, model settings, knowledge sources, integrations, sub-agents, and more.
What's new
Conversational agent creation and editing
Describe what you want to build in natural language. The builder proposes a plan, waits for your approval, then configures the agent (including instructions, model settings, knowledge, integrations, sub-agents, labels, and prompt suggestions) all at once. You can also use it to edit an existing agent: just tell it what you want to change.
Live preview as it builds
Changes stream into the agent editor in real time as the builder works, so you can see exactly what's being configured without waiting for it to finish.
Structured clarifying questions
When the builder needs more information before it can proceed, it asks focused, structured questions. You click through the options and it continues.
Visual feedback on active edits
A subtle overlay highlights the field currently being updated, so you always know what the builder is working on.
Workspace-aware suggestions
The builder has full context of your workspace. It knows which integrations you have connected, which knowledge bases are available, and which other agents exist, so it can make smart suggestions without you having to look anything up.
How to access it
The chat agent builder is available in the agent editor. When you create a new agent or open an existing draft, the builder chat appears on the left side of the editor.
Note: the preview chat used to test your agent has also moved to the left side of the editor, so both chat interfaces are now in one place.
We're excited for you to try it out!
- New interface languages: The Langdock web app is now fully available in French, Spanish, and Italian, providing a native-language experience for more users.
- Improved code canvas artifacts: Code canvas artifacts now show a live rendered preview on the card after generation, handle links correctly within the sandbox, and produce significantly better-looking output.
- Larger skill package uploads: You can now upload skill packages up to 20 MB, double the previous 10 MB limit.
- Visible errors for failed workflow trigger polls: Failed workflow trigger polls now surface as failed workflow runs with a clear error message, so you can easily identify and troubleshoot issues instead of them silently disappearing.
- More model options in Workflow Chat: Workflow Chat now supports Opus and reasoning/thinking models, including an extended-thinking toggle, giving you access to more powerful AI models directly within your workflow building experience.
- Search and filter in the Library: You can now search and filter folders and knowledge bases by name directly in the Library, making it faster to find what you need.
- Select all files in Library views: You can now select all files at once in Library views, including recent files, folders, and knowledge bases, making it faster to apply batch actions to your entire filtered file set.
- Visual understanding for images in Folders: Images uploaded to Folders can now be read and understood visually by the AI. Simply ask the model to describe or analyze an image after uploading it to a Folder.
- requiresConfirmation flag in the integrations API: The public integrations API now exposes the requiresConfirmation flag on actions, allowing API consumers to read and set whether a user confirmation is required before an action executes.
- Optional fields in action confirmation UI: You can now add and remove optional fields directly from the action confirmation UI when using integrations, giving you more flexibility to customize inputs before confirming an action.
- Custom creator display name for agents: Workspace admins can now set a custom creator display name for each agent. It appears on agent cards, the chat start screen, and agent info dialogs. This is useful when a larger team built an agent, and the name should reflect that rather than a single person. Admins can configure this in the agent settings.
Langdock for Excel
Today, we’re introducing Langdock for Excel, available on the Microsoft Marketplace. Langdock now lives directly inside Excel, so everything you need is right there in your spreadsheet.
Why we built this
Spreadsheets are where much of the work happens, but switching between Excel and a separate tool breaks the flow. Copying data back and forth, losing context, and constantly changing windows don’t scale when you’re trying to get real work done.
Langdock for Excel brings Langdock directly into Excel. You can read your workbook, write into it, and create new files with AI. Your existing Langdock integrations, agents, and skills are available without leaving your spreadsheet.
What you can do
- Ask questions about your workbook: get explanations of complex sheets, spot anomalies, or understand what a formula is doing without manually digging through cells.
- Create formulas, tables, charts, and pivot tables: describe what you need in plain language and let Langdock build it directly in your sheet.
- Clean and structure messy data: standardize names, dates, categories, or currencies before analysis.
- Summarize spreadsheet findings: turn data into clear text for reports, emails, presentations, or meetings.
- Save your work to the library: upload finished spreadsheets directly to your Langdock library so they’re always accessible when you or your team need them.
- Use Langdock features directly in Excel: knowledge bases, integrations, skills, and workflows are all available inside the plugin.
Availability
Workspace admins can enable Langdock for Excel at the bottom of the workspace integration settings. Once enabled, users can install it from the Microsoft Marketplace and start using Langdock directly in Excel.
You can check out our documentation to get started.
Claude Sonnet 5
Anthropic’s Claude Sonnet 5 is now available on Langdock! 🚀
Today, we are adding Claude Sonnet 5 to Langdock. It is Anthropic’s most agentic Sonnet model yet. It can make plans, use tools, and run autonomously at a level that previously required larger and more expensive models. Its performance is close to Opus 4.8, but at a lower price.
You can read more about the launch in Anthropic’s official announcement.
The model is hosted in the EU and available to all workspaces now.
To get you started quickly, we have automatically enabled Sonnet 5 in all workspaces where a previous Sonnet version was activated. Admins can manage model access in workspace settings.