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 project managers

Team Retrospective Follow-through Agent

Retrospective actions judged by outcomes, not activity.

AnalysisManagementIT and developmentProduct development
See how it worksGet access
Team Retrospective Follow-through Agent: what goes in, what the agent does and what you get

What it does

Retrospective actions often get marked done without anyone checking whether they solved the problem. Before each retrospective, this agent checks whether each agreed action was actually carried out, using tickets and records. Then it looks at the outcome metric linked to the original problem. If the action ran but the problem has not improved, it proposes a revised experiment with a clear measure. It prepares a learning review for the team. Policy changes and judgments about people stay with the team. Edge case: a new review checklist was adopted, but bug escape rates did not drop, so the agent proposes testing a different step.

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, continueApprovedNo 1 STARTS WHEN Retrospective approaching 2 USES A TOOL List actions from the last retrospective 3 USES A TOOL Check action execution 4 USES A TOOL Probe outcome metrics 5 CHECKS THE RESULT Did the problem improve? If not: propose a revised experiment with a clearmeasure. Back to step 4. 6 DOES Prepare the learning review 7 YOU APPROVE Team agrees on next experiments 8 RESULT Retrospective learning record
Read the steps as a list
  1. Retrospective approaching
  2. List actions from the last retrospective
  3. Check action execution
  4. Probe outcome metrics
  5. Did the problem improve?If not: propose a revised experiment with a clear measure. Back to step 4.
  6. Prepare the learning review
  7. Team agrees on next experimentsThe agent waits here for your OK.
  8. Retrospective learning record

How it decides

It distinguishes done from effective and proposes a new experiment when needed.

  • Done is not the same as effective.
  • Judge outcomes only after the agreed time.
  • Each revised experiment has one measure.

Make it yours

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

  • Outcome metric per action
  • Time allowed before judging an outcome (default 4 weeks)
  • Sources for execution evidence
  • Format of the learning review

What keeps you in control

It always asks you first

  • Team policy changes
  • Personnel judgments

Hard limits

  • No personnel judgments.

It stops when

  • Done: review prepared.

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 happensBefore the May 20 retrospective, the agent checked four actions from April. All were executed. The action to add a deploy checklist was meant to cut rollbacks, but rollbacks stayed at five per month, so the outcome check failed. It proposed a two-week trial of staged deploys with rollbacks as the measure. The team accepted it at the retrospective.

More agents for project managers

Decisions · Management

Cross-Team Dependency Resolution Agent

Dependencies unblocked based on evidence, with alternatives when they're not ready.

7 steps1 check5 jobs
Data cleanup · Management

Meeting Decision Reconciliation Agent

One consistent decision history with attributed differences.

8 steps2 checks4 jobs
Decisions · Management

Project Scope Boundary Agent

Scope changes are visible and decided, never accidental.

7 steps1 check2 jobs
Reporting · Compliance

Research Grant Deliverable Evidence Agent

Grant reports that state progress exactly as evidenced.

8 steps1 check4 jobs