Skip to content

mcs-mcp v0.21.0

Choose a tag to compare

View all tags
@github-actions github-actions released this 26 Apr 18:05
· 3 commits to main since this release
v0.21.0
1fca63a

Changes

One [start, end] range scopes every diagnostic. Set it once with
set_analysis_window, read it with get_analysis_window, and every
subsequent analysis — throughput, cycle time, flow debt, WIP, yield,
residence time, status persistence, process stability — uses the same
window. Shifting "one month back" is a single call, not ten.

Two tools use only the window's End by design:

  • analyze_work_item_age — point-in-time snapshot at End.
  • analyze_process_evolution — long-term anchor at End with a fixed
    12-month / 26-week lookback over complete buckets.

Forecasting (forecast_monte_carlo, forecast_backtest) keeps its
own engine-driven sample window. Forecast accuracy is independent of
the diagnostic lens.

The window defaults to a rolling 26 weeks ending today, lives in
memory only, and resets on board switch. Every tool response now
carries a session_window block in its context so the agent always
sees which window shaped the output.

The per-tool history_window_weeks, history_window_days, and
history_window_months parameters are removed. Set the window once;
all diagnostics follow.

Changelog

  • 72672bc feat: add SLE Adherence Trending to analyze_cycle_time
  • ae61a58 feat: add session analysis window foundation
  • 3831b16 feat: anchor analyze_process_evolution at session window End with fixed long-term lookback
  • 67671d1 feat: anchor analyze_work_item_age at session window End
  • 34ec613 feat: migrate 26w-default diagnostics to session window
  • 618ec07 feat: migrate analyze_cycle_time and analyze_yield to session window
  • 7707435 feat: migrate analyze_residence_time to session window
  • befaf20 feat: migrate analyze_status_persistence to session window
  • 1343276 refactor: post-window-migration cleanup
Assets 10
Loading