Prompt detail

Flow Delay Concentration

Location of the flow areas where time is accumulating, with investigable hypotheses rather than a root-cause verdict.

Prompt 03
Analyzed reports
PR Cycle TimeIssue Cycle TimeThroughputDaily WIPCode ReviewAging BranchesAging IssuesDeployDORA MetricsOpen PRs
Context to provide

Provide Squad-Name, a period (30 days is a useful default), and an optional repository. Say whether you want the current period compared with the previous equivalent window.

Copyable prompt
2,489 characters
ROLE Engineering flow diagnostician using MCP DevStats. OBJECTIVE Identify where time is concentrating in one squad's delivery flow and turn that evidence into bounded hypotheses and context questions. Help the user investigate root cause; do not declare root cause, prescribe fixes, assign blame, or choose priorities. ...

How the analysis runs

The analysis compares elapsed time across delivery stages, drills into no more than two evidenced concentrations, and presents alternative explanations without claiming root cause.

  1. 01Scope and inputsAlign the flow comparisonValidate Squad-Name, dates, filters, and equivalent windows so changes in PR, issue, and WIP signals can be compared safely.View detailsHide details
    Validate Squad-Name, dates, filters, and equivalent windows so changes in PR, issue, and WIP signals can be compared safely.
  2. 02Analysis and evidenceLocate the dominant time concentrationCompare total and stage cycle time, then use queue, review, aging, or deployment reports only for the one or two concentrations actually shown.View detailsHide details
    Compare total and stage cycle time, then use queue, review, aging, or deployment reports only for the one or two concentrations actually shown.
  3. 03Interpretation limitsTest bounded explanationsDescribe compatible system conditions, preserve alternative explanations, and state what elapsed-time evidence cannot prove about cause or ownership.View detailsHide details
    Describe compatible system conditions, preserve alternative explanations, and state what elapsed-time evidence cannot prove about cause or ownership.

Expected Output

Markdown · Illustrative example

Flow Delay Concentration — Squad-Name — May 2026

Where time concentrates

AreaEvidenceWhy inspectWhat it does not prove
PR review3.4 of 7.2 days; previously 2.0Largest stage increase (+70%)Reviewer behavior or lack of ownership
Issue progress11.8 days; previously 9.6Increase appears across 18 completed issuesA specific dependency or blocker

Supporting signals

  • Open PRs rose from 8 to 13 at period end.
  • Median PR size stayed near 214 changed lines.
  • Throughput changed from 36 to 34 merged PRs.

Alternative explanations

  • The evidence is compatible with a larger review queue.
  • Stable PR size leaves policy, availability, and work mix as unresolved context.

Questions to investigate root cause

  • Where did the 13 open PRs wait longest?
  • What review policy or shared dependency applied during the final two weeks?

Evidence used

  • PR Cycle Time — locate stage concentration
  • Open PRs — current queue context
  • Code Review — check size and activity

Stop guessing. Start asking.

Connect DevStats MCP to your AI assistant and turn these prompts into instant answers. Copy a prompt and go.

Get a demoStart free trial