- Home
- Approvals in the thread
Product
AI agent approval before action: a yes in the Slack thread
AI agent approval before action is how Glad works by default. Glad asks for a yes in the thread before any action that changes data or sends something, except private changes that only the requester sees, such as a draft. Reads never ask. The approval card shows exactly what will happen, the person who asked clicks Approve or Deny, and every decision is logged. It works on every plan.
That's human in the loop for AI in Slack: a named person, in the thread where the work was asked for, deciding on the exact action before it runs.
Last updated 12 October 2026
Example
- To
- marit@kiteline.example
- Subject
- Your renewal quote
- Text
- Hi Marit, as promised, here's the renewal quote for next year...
How it works
- 1
You ask in a thread.
For example: "@Glad email Kiteline's founder the renewal quote from this thread." - 2
Glad reads and prepares.
Searching, looking things up and reading a thread run without asking. Glad then prepares the action. - 3
Glad posts an approval card in the same thread.
The card shows what will be sent or changed: the recipient, every Cc and Bcc, the subject, the start of the text, and other details such as the repository or channel. Attached files show by name, type and size. The card is posted in the thread, so people in the channel see what will be sent. That's the point: the team sees it too. - 4
The right person decides.
The person who asked approves actions on their own connections. For a connection an admin shared with the team, an admin approves. Anyone else who clicks gets a private note: "You are not allowed to approve this." - 5
Glad runs exactly that action, once.
The card turns into one line, such as "Approved by Ana.", and Glad reports back in the thread. If you click Deny, nothing runs. If nobody answers, the approval expires and nothing runs.
What you can do
- Approve or Deny on the card in Slack. Neither costs a task.
- If you're an admin, or the person who asked is one, you'll also see Approve and always allow. That tool then runs without asking.
- Admins keep the always allowed list in Settings, Approval rules, with who added each tool and when. Take a tool off the list and it asks first again.
- Admins pick which tools Glad may use, app by app. Glad can't use a tool that's off.
- The Approvals page in the console keeps the record: All, Waiting and Decided, with the action (for example "Gmail: Send email"), the status, who asked, and who decided and when.
- The audit log has every request for approval, every yes or no, every expiry and every tool run.
- A
/deeprun counts as 5 tasks, and Glad asks before it starts one.
Who controls what
The person who asked approves or denies actions on their own connections, and sees their own approvals, with the details, on the Approvals page.
Admins approve actions on shared connections, set a tool to always allowed and take that back, turn tools on and off, and see every approval in the workspace. Each always-allow change is in the audit log.
Rules nobody can change:
- Glad can't approve its own action or add a tool to the always allowed list.
- A message from Glad to anyone but you, or a post in a channel, asks first every time. Always allow never applies to it. A direct message to yourself runs without asking.
- An approval covers the action on the card, once. It can't be reused, and only the first click counts. A late click gets: "This approval was already decided or has expired."
- In a channel shared with another company, only your own workspace's members can ask or approve.
- A deactivated member's decision no longer counts.
How other tools handle it
| Tool | When it asks before acting | Who approves | Admin choices per tool | Plan |
|---|---|---|---|---|
| Glad | When it asks before actingAny action that changes data or sends something, except private changes; reads never ask | Who approvesThe person who asked; an admin for a shared connection | Admin choices per toolAlways allow; tools on or off | PlanEvery plan |
| Viktor | When it asks before actingA fixed list of actions, plus any tool an admin sets to ask | Who approvesNot stated beyond "someone" | Admin choices per toolOff, Run automatically, Ask for confirmation | PlanEvery plan |
| Lindy | When it asks before actingWrites on integrations set to Require approval, in shared Slack threads | Who approvesThe person who asked, or any workspace owner or admin | Admin choices per toolAlways allow, Require approval, Don't offer | PlanEvery plan |
| Slackbot | When it asks before actingActions you set to Needs approval | Who approvesNot stated | Admin choices per toolAlways allow, Needs approval, Blocked, Custom | PlanNot stated |
| @ChatGPT | When it asks before acting"asks for your approval before taking action in connected tools" | Who approvesNot stated | Admin choices per toolNot stated | PlanBusiness, Edu, Enterprise |
Why it matters
The EU AI Act asks that high-risk AI systems can be "effectively overseen by natural persons" (Article 14), and the approval card brings that kind of oversight to the changes your team will see.
Questions
What is human in the loop for an AI agent in Slack?
In Glad, a person decides before the AI acts. The AI agent prepares the action, posts it as a card in the Slack thread, and runs it only after the right person clicks Approve.
Which actions need approval in Glad?
Any action that changes data or sends something, such as sending an email, updating a deal or filing a ticket. Private changes that only the requester sees, such as a draft, run without asking, and reads never ask. An admin can set a tool to always allowed.
Who can approve an AI action in Glad?
The person who asked, for actions on their own connections. An admin, for actions on a connection shared with the team. Anyone else gets a private note that they aren't allowed to approve.
What happens if nobody approves?
Glad lets the approval expire, and nothing runs. The run is free. Ask Glad again when you're ready.
Can an admin turn approvals off?
Glad lets an admin set a tool to "always allow" and undo it at any time. That works per tool, never as one switch for everything, and it never applies to messages Glad sends to other people or posts in a channel. Glad can't allow itself anything.
Does approving cost a task?
No. Approving and denying are free. A task counts only when Glad uses a connected tool or makes a document.
Keep reading
- Memory you can editSee every memory Glad keeps, where it came from, and edit, delete or restore it.
- Audit log, limits and rolesAn audit log, per-person task limits and admin roles on every plan, with CSV export.
- Glad vs ViktorCredits or a flat price: what a month costs, whose accounts the work runs on, and how to switch, with sources.
- Glad vs LindySeats or one workspace price: what 10 people cost, whose accounts the work runs on, and how to switch, with sources.
- Glad vs SlackbotWhat Slackbot already does in Slack, what Glad does in your other tools, and what each costs, with sources.
See it in a real task
- Sales
HubSpot
Slack
Update HubSpot deals from Slack
Say what changed on the call. Glad finds the deal, shows you the change and saves it when you approve.
See how it works - Engineering
Linear
Jira
GitHub
Slack
Create Linear, Jira and GitHub tickets from Slack threads
Glad turns the bug thread into a ticket and files it in Linear, Jira or GitHub once you say yes. Ask, and it checks for a duplicate first.
See how it works - Finance
QuickBooks
Stripe
Gmail
Slack
Find overdue invoices in Slack
Who's late in QuickBooks and by how much, as a CSV, with a Gmail reminder drafted for each. Nothing is sent until you say yes.
See how it works
Try Glad in your Slack.
50 free tasks. No card needed.
Add to Slack