Connect your tools once. Your agents take it from there.
Paste one key, limited to what the agent needs. Qoren registers the webhooks that wake your agents, keeps the key off their machines, and asks you before anything risky.
Integrations is in early access: Qoren is turning it on account by account. Already have access? Open Integrations
Wiring a tool to an agent is homework too
Every agent that reacts to a CRM, a store or a payment needs the same plumbing, and it breaks quietly. Qoren does it for every tool you connect.
- Copying a webhook URL and a signing secret into each tool, then keeping them in sync.
- Storing API keys on the agent's machine and hoping nothing reads them.
- Writing an API client for every tool, with retries and rate limits.
- Noticing that a webhook was disabled after a week of failed deliveries.
What happens when an event arrives
Ready to connect
Each tool has its own page: what agents do with it, the events that can wake them, the tools they can use, and which of those always ask first.
Pipedrive
Wake an agent when a deal, person or activity changes, and let it read and update your CRM.
9 events, 11 toolsGitHub
Wake an agent on issues, pull requests and pushes, and let it read and comment on your repositories.
9 events, 9 toolsLinear
Wake an agent when issues and comments change, and let it triage, comment and update issues.
5 events, 7 toolsStripe
Wake an agent on payments, invoices and subscriptions, and let it look up customers and charges.
10 events, 9 toolsResend
Wake an agent when an email bounces, is opened or gets a complaint, and let it check what was sent.
9 events, 3 toolsSendGrid
Wake an agent when email is delivered, bounces or is opened, and let it check sending stats.
10 events, 4 toolscal.com
Wake an agent when a meeting is booked, rescheduled or cancelled, and let it look up bookings.
4 events, 4 toolsShopify
Wake an agent when an order, customer or product changes, and let it look up your store.
8 events, 8 toolsSlack
Wake an agent when someone mentions it or posts in a channel, and let it read threads and reply.
5 events, 4 tools
Coming soon
These need sign-in through the provider rather than a key, so they arrive later. Each page says what works in the meantime.
One key, checked before it is saved
- 01Pick a tool and make its key with the least access it needs. The guide lists exactly what to tick.
- 02Paste it. Qoren checks it first: the account it belongs to, what it can do, and when it expires where the tool says.
- 03Give an agent access, read only or read and write. Nothing has access until you grant it.
- 04Pick the events that wake the agent. Qoren registers the webhooks in the tool, so there is no URL or secret to copy.
Connecting a client's account? Send a link
An agency does not need its client's key. Send a connect link instead: it shows your agency's name and the exact permissions to tick, works once, expires after 7 days and can be revoked. The client pastes the key on that page, Qoren encrypts it on arrival, and you only ever see its last four characters. A client's connection is used only by that client's agents. Connect links come with Clients, on the Ultimate, Business and Enterprise plans.
Qoren for agenciesSecurity in plain words
Keys stay in Qoren
Keys are encrypted at rest with AES-256-GCM and used only inside Qoren's own calls to the tool. They never reach an agent's machine, and after you connect only the last four characters are shown.
No access until you grant it
A connection does nothing on its own. You give each agent read only or read and write access, and only the account owner can connect, grant or revoke.
Risky actions always ask
Refunds, cancellations, archives, merges and emails to outside recipients wait for a person's approval, whatever the agent's autonomy setting.
Every call is logged
Each connect, grant and tool call, read or write, lands in the audit log with the agent, the tool and the record ids. Reads are capped at 500 records an hour per agent and connection by default.
In the docs
Step-by-step guides for connecting each tool, sending connect links to clients, and what Qoren does when a key stops working.
Frequently asked questions
What is a Qoren integration?
A tool you connect to Qoren once, with a key. Qoren then wakes your agents when something happens in that tool, and lets the agents you choose use a short list of curated tools for it, with approvals for anything risky and an audit log of every call.
Can I use Qoren Integrations today?
Integrations is in early access: Qoren is turning it on account by account. Request early access from this page and tell us which tools your agents should use. If it is already on for your account, Integrations is in the console's sidebar.
Which tools can I connect?
Pipedrive, GitHub, Linear, Stripe, Resend, SendGrid, cal.com, Shopify and Slack today. HubSpot and Gmail are coming soon.
Do my agents get my API keys?
No. Keys are encrypted in Qoren and used only inside Qoren's own calls to the tool. They never reach the agent's machine, and after you connect only the last four characters are shown.
How do agencies connect a client's tools?
An agency does not need its client's key. Send a connect link instead: it shows your agency's name and the exact permissions to tick, works once, expires after 7 days and can be revoked. The client pastes the key on that page, Qoren encrypts it on arrival, and you only ever see its last four characters. A client's connection is used only by that client's agents. Connect links come with Clients, on the Ultimate, Business and Enterprise plans.
Qoren for agenciesWhat if my tool is not on the list?
Any tool that sends signed webhooks can still wake an agent through a manual webhook trigger, and an agent can call any API with a key you keep in the vault.
Webhook triggersPaste one key. Skip the plumbing.
Integrations is in early access, turned on account by account. Request it and tell us which tools your agents should connect to.