Guide
Fibery AI Agent is a built-in assistant that can query your workspace data, answer questions, generate reports, and modify your workspace structure and content through natural language.
To open the Agent, click Fibery AI in the top left corner or press Cmd + J (Ctrl + J on Windows).
The Agent interprets your request, builds an execution plan, and asks for confirmation before making changes. For complex tasks, it breaks work into steps and may ask clarifying questions. You can see its reasoning in the Thinking section of the chat. Chat history is accessible via the History icon in the top right corner.
Ask things
First, it can query your workspace data, follow relations across databases, read rich text fields, and return answers, summaries, or formatted reports.
https://www.loom.com/share/7fc8dc4c46834fc3a7b269431aae514f?sid=b45aadd1-f976-4113-aba1-d3476ceaff46
Query your data
Ask questions about any database in your workspace. The Agent can filter, sort, and follow relations, so queries can span multiple databases.
"How many Features are currently in progress?"
"Show me all Tasks linked to Features that have no Product Area assigned."
"Which team members have vacation scheduled next month?"
Generate reports
The Agent can produce HTML and PDF reports from your data. It can also classify and group entities using semantic analysis — for example, grouping customer feedback by theme without relying on pre-existing tags.
Access to history
The Agent has access to entity history — who changed what, when, and what was discussed in comments. This is useful for catching up after time away or reviewing how a feature or project evolved.
"What changed across all Features last week? Describe each change and who made it."
"Write a narrative of this Feature's implementation timeline, including all comments and decisions."
For detailed history requests, you can guide the output format. A more specific prompt produces better results:
Prompt:
I was absent last week generate me nice overview about changes in all the features for last week. I want to be informed very well about this to catch up with work. I am more interested in text and narrative description than just tables and charts. Describe every changed feature and what was changed. Separately create a section with newly added features.
Include:
- What exactly was changed and by whom narrative
- What exactly was discussed and what decisions were made related to the feature
- Include correct links to features
Do things
Fibery AI can do things (manage spaces, databases, fields, views, formulas, entities) in Architect mode.
Workspace Schema Management
Instead of clicking through menus to set up your workspace structure, just describe what you need in plain language. The AI Agent can handle most schema changes for you.
Organize your workspace. Create new Spaces or remove ones you no longer need.
Set up and modify databases. Need a new database for tracking customer requests? Want to add a priority field to an existing one? The Agent can create, update, or delete databases and their fields: text, numbers, dates, single-selects, rich text, and more. It can also add or reorder options within single-select fields.
Connect your data. Tell the Agent what you want to connect (e.g., "each Feature should belong to one Product Area, but can have multiple Tags") and it will figure out the right relation type: one-to-one, one-to-many, or many-to-many.
Restructure without starting over. Realized a field belongs in a different database? Want to split a database that's gotten too bloated? The Agent can move and refactor fields and relations between databases while preserving your existing data.
If you're not sure how to structure something, ask the Agent to suggest an approach before building it. For example: "I need to track bugs, features, and customer requests — how should I organize this?" It will outline options so you stay in control of the design.
Data Operations
The AI Agent can read and modify data across your workspace. This includes both structured fields and rich text content.
Querying data. The Agent can search, filter, sort, and analyze entities across any database. It follows relations, so you can ask questions that span multiple databases (e.g., "show me all Tasks linked to Features that have no Product Area assigned").
Creating and editing entities. You can ask the Agent to create, update, or delete entities — individually or in batch. This covers all field types: text, numbers, dates, single-selects, relations, and so on.
Bulk operations. The Agent can apply changes across many entities at once: assign owners, update statuses, set relations, move entities between groups. Describe the condition and the change, and it will handle the rest. Examples: "assign all unassigned Tasks in Sprint 12 to Jane" or "move incomplete Tasks from Release 1 to Release 2."
Data migration and transformation. The Agent can move data between fields, between databases, or convert it from one format to another. This is useful when restructuring your workspace — for instance, converting a text field into a single-select while preserving existing values.
Rich text (document) fields. The Agent can read and write document content in markdown, HTML, or plain text. It can also copy or transform document content between entities. For Document Views, it supports both full replacement and appending to existing content.
Views & Visualization
AI agent can create and modify Views, including Table, Board, Calendar, Timeline, Gantt, Feed, Map, Form. AI can apply filters and sorting (but not color coding yet) as well as View-specific settings (e.g., Milestones on Timeline).
Report Views
AI agent can create and modify Reports. Reports are relatively hard to create, especially when you want to add some custom calculations. But with Fibery AI you can create new report, set it into the context and iterate till it is good enough.
NOTE: Fibery AI does not support history/cumulative reports yet.
❌ DISABLED SO FAR: Create automation rules & buttons
We temporarily disabled this feature until we make it better, sorry.
Fibery AI can create automation rules or buttons for your databases. Ask it to create a Rule and iterate to make changes to the rule.
Known issues:
After modification Rule is not reloaded, so you have to reload it manually to see changes…
Sometimes filters in triggers may be too complex for UI to parse them well, but they may still work fine
Sometimes AI uses Script for action extensively
Integrations & HTTP
AI Integrations agent can create custom integrations for you. Check Creating integrations using AI guide. In short, it can
Create new Fibery integration connectors for any HTTP/HTTPS API (e.g. Linear, Stripe, internal REST APIs)
Sync specific entities (Issues, Projects, Tickets, etc.) with custom filters
Inspect and edit connector code directly in the browser
Troubleshoot and fix errors with AI assistance
Optionally enable auto-fix, so the AI can repair broken connectors when APIs change
If you don't need full integration, AI Agent can just make HTTP GET/POST requests to external APIs for integrations or data enrichment. So you can ask it to fetch data from some external data source and put this data into new Database. For example, here is a prompt that will get all countries from public API and create a new database with all the countries
Prompt:
Fetch countries from here https://restcountries.com/v3.1/all?fields=name,cca2,cca3,capital,region,flags and put them into Countries database
And here is the result in Fbery:
Some use cases ideas
We ourselves just started to comprehend the enormous power of this tool… Some trivial things it can do:
Add sample data to databases fast for demonstration or testing
Merge two databases (video)
Duplicate database with or without data (video)
Split a database (video)
Move or copy a field from one database to another (and keep the content) (video)
Convert text field to single select and keep all values
Convert single select field to a database (video)
Add new Tags database, add many-to-many relation between Tags and Content databases, create and set relevant Tags for all Content entities (video)
Batch update entities based on some fuzzy or exact conditions, like "find all features without Product Area relation and set a relation to some product area that fits best" or "move not-completed tasks from Release 1 to Release 2" (video)
Remove old, unused tags from all articles
Normalise inconsistent field values
How Fibery AI Agent works
Fibery AI Agent first interprets what you're asking and create plans how execute your request. It often asks for confirmation and then writes code to execute the necessary queries and commands. For complex requests, it breaks tasks into smaller steps and may ask clarifying questions along the way.
AI Agent has quite many specific tools. If you want to get deep, ask it "what tools do you have?" and check for yourself what you can use. Note that you can always guide the agent to use specific tools, so do it for fine tuning your prompts.
You can see what Smart Agent is doing in Thinking section.
Limitations
AI Agent knows nothing about Images, it only knows full databases content and Documents text
AI Agent can't do the following things (yet):
There is no easy undo. In general all schema changes and data deletions are reversible from Audit History, but data modification will be hard to restore.
Can't setup and read permissions (on any level - entity, database, space)
Cant't work with whiteboards & color-code ANY views
Can't embed views in rich-text fields or document views
Can't debug automation scripts
Can't setup existing integrations and configure them
Can't share anything (apart from form) to the web
Can't comment from user's perspective, as well as add reactions to comments
Can't export anything (for example, view in CSV format)
Can't manage workspace users (invite, deactivate them, change roles etc)
Can't read/change notification settings
Can't read/change user settings (such as theme, workspace name, timezone, enable/disable experimental features)
Can't see billing information & invoice history
FAQ
Is there a free trial of Fibery AI? How much does it cost?
It is free end included into every plan. There are limited number of queries you can run every month, for Free it is 20 queries per month, for Standard it is 50 queries + 10 for every paid user, for Pro it is 100 + 25 and for Enterprise it is 200 + 50.
Check out our pricing page to see the limits.
Can I turn Fibery AI off for the whole account?
Yes, go to Settings → AI.
What are the privacy implications?
If you don’t use this feature, the privacy situation is the same as before.
We don’t use the prompt's text as training data. Check OpenAI Terms. According to the latest version they don't use your data as well:
We do not use Content that you provide to or receive from our API (“API Content”) to develop or improve our Services.
For additional information specific to Fibery AI, see the Fibery AI Additional Terms.
Which models does Fibery use?
Fibery uses different models for different use cases from Anthropic, Google and OpenAI.