Weekly Flow Board
Status: Live.
Your muse
A kanban board with drag and drop between To do, Doing and Done columns, local persistence so cards and their order survive a reload, and a Spark-written summary of the week from the cards moved to Done.
Checks passed
The plan: Weekly Flow Board
Drag tasks to Done and get a clear weekly recap.
- Add cards to To do with a title and optional note
- Drag cards between To do, Doing and Done and reorder within columns
- Reload safe because board saves in the browser
- Press Summarize week to send Done cards to Spark
See the full plan
Who it's for
Individuals and small teams tracking weekly tasks.
How it works
- Add cards to To do with a title and optional note.
- Drag cards between To do, Doing and Done and reorder within columns.
- Reload safe because board saves in the browser.
- Press Summarize week to send Done cards to Spark.
- Read and copy the weekly summary.
Screens
- Task Board. Three columns for To do, Doing and Done with add, edit, delete and drag and drop.
- Week Summary. Panel that shows the Spark written summary of Done cards with copy and regenerate.
What goes in
- Text: Card title with optional note, under 500 characters per card.
What comes out
- Text: Short weekly summary paragraph with wins and open items, ready to copy.
Models
Model Used for Why Muse Spark 1.3 Write a weekly summary from the Done cards Reads card titles and notes and returns short summary text as JSON or plain text. Limits
- Board data stays in this browser localStorage only. It does not sync across devices.
- Clearing browser data deletes cards and order.
- Summary needs at least one card in Done. Empty Done gives a prompt to add work first.
- Summary is generated text and may miss context not written on cards.
From your inspo
The person gave a text only note asking for a kanban board with drag and drop, local persistence, and a weekly summary of Done cards.
Takes
- The kanban job with To do, Doing and Done columns
- Drag and drop flow and order persistence
- Weekly summary written from Done cards
Cost per run: about $0.003 (est.). One Spark call with about 800 input tokens at $1.25 per million and about 350 output tokens at $4.25 per million. Input about $0.001 plus output about $0.0015. Total about $0.003.
Recorded demo Done.
npm run verifyexits 0: typecheck, build, start, console, smoke (5 passed), and manifest all green.