Guide
When we talk about Fibery in context of integrations, I think it's better to think of it as your central hub. Integrations let you pull data from external tools into Fibery, so that information from the tools you can't replace (🥲) is still connected to the work you manage here.
The sweet spot for integrations is when you already run meaningful processes in Fibery and need data from other tools to make better decisions in those processes. For example: your product work lives in Fibery, and you want GitHub Pull Requests linked to your features to track delivery. Or your CRM is in Fibery, and you want customer emails synced so your team has the full picture without switching tabs.
Fibery isn't a great fit as a pure integration hub. If you're looking to aggregate data from a dozen tools in one place without running any core process in Fibery itself - MCP-connected AI assistants will do a better job these days. Integrations shine when they enrich work that already lives in Fibery.
This guide explains how integrations work, what kinds exist, and what you can and can't do with synced data.
How integrations work
As you may know, Fibery's data model is flexible by design - but the tools you integrate with might not be, ha, losers. Take Jira as an example. It has a strict data structure - everything new becomes an Issue type, and if your process doesn't map cleanly onto that, you have to force it to fit.
Fibery's own data model is the opposite of that - flexible by design (that's why you're here, right?…). But we can't change how other tools store their data. So instead, when you connect an external tool, Fibery pulls in its data as a set of connected Databases that mirror the structure from that tool (Issues, Contacts, Pull Requests, and so on). Those databases are read-only as far as the synced fields go - the source tool still owns that data.
What you can do is build on top of them: add your own fields, create Relations to other Fibery databases, build Views and Charts, write Formulas. Integration data becomes a first-class citizen. :happy-dog:
For example here we see that the Intercom integration database Company is connected to other databases from the same integration (Contact and Tag) but also to the Workspace database from our CRM.
Data is synced on a schedule you configure (every 60, 120, or 240 minutes, daily) or manually. Some integrations also support webhooks for near-instant updates when something changes in the source. And you can always force a manual update anytime. Humans are still powerful! :excited-llama:
Choosing which fields to sync
When you set up an integration, you choose which fields from the external tool to bring into Fibery. You can change this later via Edit Fields to Sync in the integration settings.
Syncing fewer fields keeps your databases cleaner and your syncs faster - external tools often expose plenty of fields you don't actually need.
Fibery supports a maximum of 234 fields per integration database. So, for tools with large schemas (those that have lots of fields like HubSpot), it's worth deciding what you really need include.
Note: When new fields appear in a source tool, they aren't enabled automatically - you'll need to add them in the integration's Fields to sync configuration. This is by design: auto-enabling unknown fields could introduce unexpected data or break existing sync logic. Once you add a field there, it will start syncing on the next run. Enabling new fields won't affect fields already syncing.
Types of integrations
There are three ways to connect an external tool to Fibery:
Template integrations
These are pre-built connectors for popular tools we liked - GitHub, GitLab, Jira, HubSpot, Intercom, Linear, and others. You can find them in the Templates gallery or through Settings → Integrations. Authenticate your account, pick what to sync, and you're done. :mario-luigi:
Build a custom connector with AI
Our list of ready-made integrations isn't very long. And there's a high chance that the tool you want to connect with isn't listed. Or you simply don't like how our integration was built. Maybe you came to our no-code tool hoping not to code - then you can hand over the job to AI.
So, to build a custom connector, you describe what you want in plain language (e.g. "sync Issues and Projects from our Linear account, auth via API key"), and the AI agent writes, tests, and saves the connector code for you. It supports API keys and basic auth; OAuth is not supported. Once saved, the connector appears in your integration gallery like any other. See Creating integrations using AI.
AI is a tool, not a magic wand. The agent handles a lot, but it's not perfect - connectors can have bugs, edge cases, or drift when an external API changes. If something isn't syncing as expected, being able to read and debug the generated script will save you a lot of time. You don't need to write it from scratch, but knowing how to inspect and adjust it helps.
Build a custom connector on your own
Custom apps are for developers who are ready to struggle with our API 🙂. You implement and host your own HTTP service that follows Fibery's integration API spec, then register its URL in Fibery. This approach supports the full range of configuration options (yep, including OAuth!). For more details, check out the developer documentation.
Why synced data is read-only and what about external actions
Integration sync is one-way.
Data moves from the external tool into Fibery, and not the other way around. The source tool owns its data, and we don't want to have different information in different tools. Synced fields are read-only as a result - you can't edit a GitHub Issue's title from Fibery and have it update in GitHub.
There are still a few exceptions. Some fields in integration databases are editable - for example, rich text fields in Intercom or Discourse conversations. This is intentional: these integrations are often used as feedback sources, and being able to highlight and extract insights from that text is a core part of the workflow. If you're curious how that works, see more details here.
This doesn't mean you're limited to just reading. There are two ways to act:
Built-in actions via Automations. For GitHub, GitLab, Jira, Slack and Email, you will find native actions you can trigger from Automation rules - for example, creating a Merge Request in GitLab when a Story moves to a certain state, or posting a comment on a Jira issue. These appear as action steps inside Automations.
Custom actions. For everything else, you can use the Send Web Request action in Automations to call any external API, optionally with JavaScript for more complex logic. You can also connect external automation platforms like Make or n8n to trigger actions from Fibery events.
Who can add integrations
Currently, only workspace Admins can set up integrations. This is a known limitation - the plan is to extend this to other roles in the future. (please let us know if you need that!)
Once an integration is set up by an Admin, all workspace members can see and work with the synced data according to their Space permissions.
FAQ
How do I change the account credentials that an integration uses?
You don't need to reinstall the integration. Open the integration's settings (click the integration icon in the sidebar → … → Configure), and reconnect or swap the account there.
How do I remove an integration?
Two options are available in the integration settings:
Disconnect and Migrate - removes the integration connection but converts the synced Databases into regular native Fibery databases. Your data stays, but it will no longer sync.
Remove and Delete - permanently deletes the integration and all its data.
What happens to synced data if the account that set up the integration loses access?
The sync will stop working. If a different user reconnects, Fibery will respect the new user's permissions in the source tool — any records that aren't accessible may be marked as Deleted in App. So, it's worth setting up integrations using a dedicated service account with stable permissions, or reconnecting before the original admin loses access.
Can I add my own fields to an integration database?
Yes. You can add custom fields and Relations to any integration database. These are native Fibery fields and won't be affected by sync.