Report Generation on Autopilot for Azure DevOps Users
At the end of every sprint, an agent reads Azure Boards, pull requests and pipeline runs, then sends stakeholders a plain-English account of what was committed, what shipped and what slipped.
Sprint 57 closed at 5:00 PM. Report built from 41 work items and 12 pipeline runs:
How can I automate sprint reports from Azure DevOps?
Connect an agent to your Azure DevOps project and tell it when your sprints end and who reads the report.
- 01
Trigger fires
On the iteration's end date it queries the sprint's work items, reads their history to separate what was committed at planning from what was added later, and matches completed stories to merged pull requests and production pipeline runs.
- 02
Compare iteration scope at planning with scope at close
It then emails stakeholders a short summary of what shipped, what carried over and which Priority 1 bugs are still open, and publishes the full breakdown to a wiki page.
- 03
Match done stories to merged pull requests and releases
Stories closed with no linked code, failed releases and unestimated items wait for the team lead instead of skewing the totals.
- 04
List carry-over, mid-sprint additions and open Priority 1 bugs
- 05
You approve
Anything under your confidence bar waits for a human.
How you tell it what to do
Built in plain English.
You write the rule the way you'd describe it to a teammate. The agent reads the rule, breaks it into the actions it'll take, and confirms the apps it'll touch — before it does anything.
- 1Compare iteration scope at planning with scope at close
- 2Match done stories to merged pull requests and releases
- 3List carry-over, mid-sprint additions and open Priority 1 bugs
- 4Email stakeholders and publish the full wiki page
How it connects
Connect Azure DevOps. The agent does the rest.
Claude and ChatGPT are already running on our side. You connect Azure DevOps with one click, and report generation runs inside it.
Claude and ChatGPT run on our keys. Nothing for you to configure.
- Azure DevOpsConnect
- Microsoft_outlookConnect
Runs on your data, in your apps.
Nothing to deploy. Nothing to maintain.
Actions
What Azure DevOps + DeskFerry can do
Real Azure DevOps actions your AI agent can perform automatically — no manual work required.
Provision cloud resources
Spin up servers, databases, or containers in Azure DevOps when deployment pipelines or scaling rules are triggered.
Monitor system health
Watch CPU, memory, and network metrics in Azure DevOps and trigger alerts when thresholds are breached.
Scale resources automatically
Adjust compute capacity in Azure DevOps based on traffic patterns, queue depth, or custom scaling policies.
Rotate secrets and credentials
Automatically rotate API keys, database passwords, and certificates in Azure DevOps on a defined schedule.
Deploy application updates
Trigger rolling deployments in Azure DevOps when new container images or build artifacts are available.
Manage DNS records
Create, update, or remove DNS entries in Azure DevOps as part of deployment or domain management workflows.
Collect and forward logs
Stream application and infrastructure logs from Azure DevOps to centralized logging and analysis platforms.
Enforce security policies
Audit resource configurations in Azure DevOps against compliance rules and remediate violations automatically.
Story is Closed but has no linked pull request, commit or build. Count it as shipped?
Production stage failed Thursday and passed on re-run Friday. Report both runs or only the successful one?
No story points on #20442, #20447 and #20451. Point totals are shown without them for now.
Human in the loop
Approve before it sends.
Every draft lands in a review queue. You approve, edit, or reject — the agent never acts on its own unless you explicitly turn that on for a workflow you trust.
Governance
Every action, with the reasoning attached.
Each step the agent takes is logged with what it did, why it did it, and which app it touched. Audit-ready, so security and compliance can sign off without backfilling.
- Lena5:14 PM
Replied asking for the report to go to exec staff each sprint.
- Agent5:06 PM
Emailed the Sprint 57 report to 2 recipients and published wiki page Sprint Reviews/57.
- Agent5:05 PM
Left #20430 out of the shipped count.
Reason: Its state is Closed but nothing links it to a pull request, build or release, so the report can't show it reached users.
- Agent5:03 PM
Linked 9 completed stories to 14 merged pull requests and release 4.18.
- Agent5:01 PM
Found 13 stories committed at planning and 2 bugs added after day 3 of the sprint.
How it works
Get started in three steps
Step 01
Connect Azure DevOps and your mailbox
Authorize the project so the agent can read Boards, pull requests in Repos and Pipelines runs, plus the Outlook mailbox or channel the report goes out on.
Step 02
Describe the report stakeholders want
Name the team's iteration path, what counts as done (Closed, or Closed with a merged PR), and who gets the short email versus the full wiki page.
Step 03
Rule on the edge cases once
Stories closed with no code, failed releases and unestimated items land in your queue. Your answer becomes part of the rule for the next sprint.
Start automating Report Generation for Azure DevOps
7-day free trial. Works with the tools you already use.
FAQ
Frequently asked questions
Why not just share the Azure DevOps burndown and Analytics views?
Those suit the team, and the agent reads the same data. The report is for people who never open Azure DevOps: it says in sentences what shipped, what was added after planning, what slipped and why, and which release carried each story, instead of a chart someone has to explain in a meeting.
How does it know what was committed when the sprint started?
It reads each work item's revision history, not just its current iteration. Items in the iteration when planning ended count as committed, items moved in later are listed as added, and items moved out are listed as descoped. That is how the report separates a scope change from a missed commitment.
Can it report on releases, not just work items?
Yes. It reads runs for the pipeline stages you name, such as a production deployment stage, and ties them back to the pull requests and work items they carried. A story that is Closed but never went through a production run is reported as done but not yet released.
Does it work across several teams with different sprint calendars?
Give it each team's area path and iteration path. Every team gets its own report on its own end date, and an optional combined page only includes sprints that have actually closed, so a director never sees one team's half-finished sprint next to another's final numbers.
What access does it need in our Azure DevOps organization?
Read access to Boards, Repos and Pipelines in the project is enough for reporting, plus write access to the wiki if you want the report published there. It doesn't need to change any work item for this job, and every query and wiki edit is recorded in the audit log.
Explore more