Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for scrum masters

Cross-Team Dependency Tracker Agent

Catch slipping cross-team dependencies early enough to re-plan the sprint

Risk managementMonitoringProject trackingIT and developmentProduct developmentManagement
See how it worksGet access
Cross-Team Dependency Tracker Agent: what goes in, what the agent does and what you get

What it does

Many sprint misses start with a dependency nobody tracked: an API from another team, a design review, a data export. The scrum master finds out on day eight. This agent reads the sprint backlog at planning and finds every ticket linked to or blocked by work in another team's board. For each link it checks whether that work is planned in the other team's current sprint and who owns it. Twice a week it rechecks status and dates. If a dependency moves to a later sprint or loses its owner, it flags the risk to the sprint goal and drafts a message to the other team's scrum master. It does not send the message. The scrum master decides whether to escalate, re-plan or swap scope. Edge case: links to closed tickets that were reopened are treated as new dependencies.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Sprint planning or twice-weekly check 2 USES A TOOL Read the sprint backlog and ticket links 3 DOES List tickets that depend on another team's work 4 USES A TOOL Look up each linked ticket on the other team's board 5 CHECKS THE RESULT Is each linked ticket planned, owned and due beforeit is needed? If not: mark it at risk and look for an alternativeowner or date. Back to step 4. 6 DOES Score the impact on the sprint goal 7 CHECKS THE RESULT Has the risk changed since the last check? If not: keep the existing flag without a new alert. Backto step 5. 8 DOES Draft a message to the other team's scrum master 9 YOU APPROVE Scrum master decides to send, escalate or re-plan 10 RESULT Dependency status table updated
Read the steps as a list
  1. Sprint planning or twice-weekly check
  2. Read the sprint backlog and ticket links
  3. List tickets that depend on another team's work
  4. Look up each linked ticket on the other team's board
  5. Is each linked ticket planned, owned and due before it is needed?If not: mark it at risk and look for an alternative owner or date. Back to step 4.
  6. Score the impact on the sprint goal
  7. Has the risk changed since the last check?If not: keep the existing flag without a new alert. Back to step 5.
  8. Draft a message to the other team's scrum master
  9. Scrum master decides to send, escalate or re-planThe agent waits here for your OK.
  10. Dependency status table updated

How it decides

A dependency is at risk when the linked ticket is not in the other team's current sprint, has no owner, or its due date is after the date the team needs it.

  • At risk if the linked ticket is not in the other team's current sprint
  • At risk if due later than two days before the need date
  • Alert only on changes, not on unchanged risks
  • Rank risks to sprint goal tickets first

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Check days (default Tuesday and Thursday)
  • Buffer before need date (default 2 workdays)
  • Teams to track (default all linked teams)
  • Where the status table lives (default sprint wiki page)

What keeps you in control

It always asks you first

  • Sending any message to another team
  • Removing scope from the sprint

Hard limits

  • Never edits other teams' tickets
  • Never sends messages without approval

It stops when

  • Done: every dependency confirmed or re-planned
  • Stop: other team's board not accessible

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensAt planning on 9 September the agent found 3 dependencies. The payments API ticket was in the platform team's backlog, not their sprint, so it was marked at risk. On Thursday it was still unplanned, now 5 days before need. The agent drafted a note to the platform scrum master. The scrum master sent it and swapped one story out of the sprint.

More agents for scrum masters

Analysis · Management

Team Retrospective Follow-through Agent

Retrospective actions judged by outcomes, not activity.

8 steps1 check3 jobs
Quality control · Compliance

Acceptance Criteria Test Mapping Agent

Make sure every accepted story has evidence for each of its acceptance criteria

11 steps1 check2 jobs
Risk management · Monitoring

Roadmap Commitment Risk Watch Agent

Warn early about at-risk commitments with options the owner can act on

11 steps1 check5 jobs
Risk management · Monitoring

Mid-Sprint Goal Health Agent

Spot a failing sprint goal while there is still time to act

11 steps2 checks3 jobs