| name | weekly-plan |
| description | Turn an unstructured backlog into a weekly plan of 3 features to build, 3 bugs to fix, and 3 users to talk to, using the 3-3-3 method. Use at the start of each work week, especially when the founder has more ideas than a week can hold. |
Weekly Plan (3-3-3 method)
There's no sprint board here, only an unbounded backlog and a founder who
will happily skip talking to users in favor of building one more thing.
3-3-3 caps scope on purpose and makes user contact non-optional, on
purpose.
What to do
-
Cap Build at 3. If the founder lists more, push back — "which 3
ship first?" Check each item against PRD.md's non-goals list and the
## Status checkboxes for what's already in flight before adding
anything new.
-
Cap Fix at 3. Pull from known bugs: ask the founder directly, check
recent /debug-seb sessions, or scan CHANGELOG.md's ### Fixed
entries. Rank by actual user impact, not founder annoyance.
-
Require 3 real user contacts — don't let this slot go empty. This
is the one founders skip first under build-momentum pressure. State
explicitly why it stays: as /validate-demand already established,
distribution and demand are the bottleneck here, not speed of building.
-
Write the plan using the WEEK OF / BUILD / FIX / LEARN / DONE WHEN
shape, with a concrete "done when" condition — not just a task list
with no definition of finished.
-
Reconcile at week's end. Compare what got done against what was
planned. Carry unfinished items forward explicitly rather than letting
the backlog grow silently in the background.
Anti-patterns to avoid
- Letting "3 features" quietly become 6 by splitting one feature into
many sub-tasks.
- Substituting dashboard-watching for real user conversations in the
Learn column.
- Running this plan internally without reading it back to the founder for
buy-in before the week starts.
