Troopr AI Scrum Master is a Slack app from Troopr Labs that drafts async standups, runs retros, and runs planning poker using each person's real Jira, GitHub, and Slack activity instead of a manual "what did you do yesterday" form. You install it from the Slack App Directory, connect Jira, and it starts writing the first draft of your standup for you.
Short answer: Troopr AI Scrum Master installs from the Slack App Directory, connects to Jira (and optionally GitHub), and drafts standups, retros, and planning poker sessions from real activity instead of manual updates. It's free forever for up to 3 seats with all features included; paid seats start at $6/user/month billed annually, or $8 month-to-month.
By Vishal Swami, Founder & Lead AI Reviewer, AISagely. Last updated August 7, 2026.
I connected Troopr to a small Jira project and a test Slack workspace to see whether "the standup writes itself" was real or marketing copy. In my testing, it mostly held up — the AI draft correctly picked out the Jira tickets I'd moved and the branch I'd pushed that morning, though I still had to add context it couldn't infer, like why a ticket was blocked. Here's the setup, what each ceremony actually does, current pricing, and where I'd stop trusting it to run unsupervised.
What you'll need
A Slack workspace where you can install apps (or an admin who can approve the install for you), and a Jira project if you want the AI-drafted standups and reports — Troopr works without Jira, but the "writes itself from real activity" feature needs something to read activity from. GitHub is optional and adds commit and PR context to the same draft. No credit card is required to start: Troopr's free tier covers up to 3 seats with every feature turned on, and the Slack listing also offers a fully-featured trial for the whole team before anyone has to pick a paid seat.
Step-by-step: setting up Troopr AI Scrum Master
1. Install Troopr from the Slack App Directory
Search "Troopr" in Slack's App Directory, or install the "AI Scrum Master: Slack Standup, Retro, Poker for Jira" listing directly from the Atlassian Marketplace, and approve the OAuth permissions. Troopr says it respects your existing Jira permissions rather than granting itself broader access, which matters if some projects are restricted to specific teams.
2. Connect Jira (and GitHub if you use it)
From the Troopr web app, link your Jira site and pick which projects it should read. Add GitHub next if you want commits and pull requests folded into the same standup draft — this is what lets Troopr say "closed 3 tickets, opened 1 PR" instead of just listing ticket titles.
3. Pick a channel and set the standup schedule
Choose a Slack channel, set the time zone(s) your team spans, and pick a schedule. Troopr supports multiple timezone prompts in one channel, so a distributed team doesn't get pinged at 3 a.m. in one region to hit a deadline in another.
4. Let the AI draft the first standup, then edit it
At the scheduled time, Troopr posts a thread with a draft standup update per person, pulled from Jira and GitHub activity. Teammates confirm, edit, or add blockers before it's finalized. When I tested this, the draft nailed the "what I did" section from ticket movement but left blockers blank, which is correct — that's not something activity data can guess.
5. Run retros and planning poker from the same app
Use the built-in retro template to collect what went well, what didn't, and what to change, and have Troopr turn agreed actions into tracked Jira issues. For estimation, start a Planning Poker session tied to a specific Jira issue; votes are anonymous until everyone's in, then Troopr reveals them together.
6. Check the automated reports
Troopr schedules Jira reports — burndown and velocity, plus nudges for tickets that have gone stale — straight to your Slack channel. I set a weekly burndown report and it posted on schedule without needing a re-check, which is the kind of unglamorous reliability that actually saves time.
Example prompts you can copy
Troopr isn't a chat-first AI tool, so most of the "prompting" happens through short setup instructions and the answers you type into its check-in questions rather than open-ended prompts. These are close to what worked in my testing:
- Standup question customization: "Ask each person for blockers first, then yesterday's work, then today's plan — skip anyone on PTO automatically."
- Retro prompt: "What's one thing from this sprint we should stop doing, and what would you replace it with?"
- Jira issue creation from a thread: "Turn this thread into a bug ticket in the Payments project, priority high, and assign it to whoever's mentioned last."
- Report nudge: "Flag any ticket that's been In Progress for more than 5 days without a comment."
- Mood check-in: "On a scale of 1–5, how manageable does this week's workload feel, and why?"
The pattern that worked best: give Troopr a rule tied to a real field (priority, days-in-status, PTO calendar) instead of a vague instruction, since it's reading structured data, not guessing intent from prose.
Common mistakes to avoid
The mistake I made first was assuming the AI-drafted standup needed no review — it's a strong first draft, not a finished one, and skipping the edit step means blockers and context never make it into the record. Second, connecting every Jira project at once instead of starting with the one team that actually wants this; a noisy rollout across a company that hasn't agreed to it is how these bots get muted. Third, ignoring the seat math: Standard pricing is per active seat, and each seat only unlocks 2 more synced channels and 2 more scheduled reports, pooled workspace-wide — a team running standups in five channels needs more seats than headcount alone would suggest. Fourth, treating Planning Poker results as final without discussing outliers; the anonymous reveal is meant to start a conversation about why estimates differ, not to auto-average them. Fifth, forgetting that readers (people who just view reports without ever checking in) are free — you don't need to buy a seat for someone who only wants to see the burndown chart.
Troopr AI Scrum Master pricing
| Plan | Price | Seats | Channels & reports |
|---|---|---|---|
| Free | $0 forever | Up to 3 seats | 3 synced channels + 3 scheduled reports total, all features included |
| Standard | $6/seat/mo (annual) or $8/seat/mo (monthly) | Unlimited seats, pay per active user; readers free | 2 synced channels + 2 scheduled reports per seat, pooled workspace-wide |
| Enterprise | Custom, annual contract | Unlimited, volume seat pricing | Unlimited channels & reports, plus Slack Grid, white-labeling, self-hosted option |
Prices and limits came straight from Troopr’s pricing page, checked directly on August 7, 2026, since this is an area vendors change without much notice. Every plan, including Free, includes the full feature set — AI issue creation, standup summaries, bidirectional Jira sync, and analytics — so upgrading buys more channels and reports, not more features. Enterprise adds the things a security or IT team would actually ask for: a self-hosted option, Slack Grid support for multi-workspace orgs, and a dedicated success manager.
Tools that make this easier
Troopr is one piece of an AI-assisted engineering workflow, not the whole thing, and it's worth knowing what pairs well with it. If your team is also evaluating AI in the editor itself, my AI coding assistant guide and best AI tool for code roundup cover the tools that write and review the code Troopr is reporting on — how to use Cursor AI is a good next read if GitHub activity is already flowing into your standups. Smaller teams deciding whether Troopr's paid tiers are worth it should see best AI tool for small business for how it stacks up against other lean tooling. Teams that manage project docs and notes outside Jira might prefer pairing Troopr's Slack reporting with how to use Notion AI for the write-up side. And if you're skeptical that any of this actually saves meaningful time, AI productivity gains are closer to 10% than 10x is a useful reality check before you build a workflow around the promise instead of the measured result.
My take
Troopr earns its place for teams that already live in Jira and Slack and are tired of standups turning into copy-pasted status text — reading real activity instead of trusting self-reported updates is a genuinely different approach, not just a chatbot bolted onto a form. It's not a scrum master replacement in the people-management sense; it doesn't run the conversation about why a sprint slipped, it just gives you cleaner input to have that conversation with. Start on the free plan with one real team before rolling it out further, since the seat-and-channel math on Standard rewards a focused pilot over a company-wide install on day one.
Frequently Asked Questions
Is Troopr AI Scrum Master free?
Yes, permanently, for up to 3 seats with 3 synced channels and 3 scheduled reports, and every feature unlocked — there's no feature gate on the Free plan. Paid seats start at $6/user/month billed annually once a team outgrows that.
How long does it take to set up Troopr AI Scrum Master?
Installing the Slack app and connecting one Jira project takes about 10–15 minutes in my testing. Getting the standup questions, schedule, and time zones tuned to how your team actually works takes a first week of adjustment before the drafts stop needing much editing.
What is the easiest way to try Troopr AI Scrum Master?
Install it in Slack, connect a single active Jira project, and run it in one channel for a sprint before deciding whether to expand it. The free plan's 3-seat limit is enough to pilot it with a small team without any billing commitment.
Does Troopr AI Scrum Master work without Jira?
Yes, but you lose the main draw — standups drafted from real ticket activity. Without Jira or GitHub connected, teammates are back to typing manual updates, just inside a scheduled Slack thread instead of a form.
Is Troopr AI Scrum Master secure enough for a company Jira instance?
Troopr states it's SOC 2 Type II certified, doesn't train models on customer data, and enforces your existing Jira permissions rather than granting itself broader access. Enterprise plans add a security assessment and a staging environment if your team needs to validate that before a full rollout.