Agile Plan Sprint

When the user asks to plan a sprint — select which backlog items will become stories in the current sprint. Requires a mission to be set on the sprint document; if none exists, establish it with the user before scoring. For general backlog housekeeping without a sprint-selection goal, use backlog-refiner instead.

1. Read the current sprint and its mission:

   ./projects/ores.compass/compass.sh list --type sprint --sort path | tail -3
   

   Open sprint.org and read * Mission and * Goals. If the mission placeholder is still unfilled, stop and ask the user for a one-sentence mission before proceeding — goal-less planning produces low-signal rankings.

2. Understand what is already committed. Read the * Stories table in sprint.org to know which stories are already STARTED or BACKLOG — these are not candidates for selection; they are already planned.

3. List the next-bucket captures:

   ./projects/ores.compass/compass.sh list --type capture --sort path | grep next
   

   Read each capture's title and description. Ignore captures that are already DONE or ABANDONED.

4. Score each candidate against the sprint goals using three criteria — present results as a ranked table:

   Criterion	Question
   Fit	Does this directly advance the sprint mission?
   Size	Can it plausibly fit in one sprint as a single story (one PR or a small set)?
   Dependency	Does it depend on in-flight work that isn't done yet?

   High-fit, right-sized, unblocked items rank first.

5. Present the ranked list to the user and agree on which to promote. Do not promote without explicit confirmation — sprint scope is the user's decision.

6. Promote each confirmed capture to a story:

   a. Scaffold the story doc in the sprint folder, preserving the capture's :ID::

   ./projects/ores.compass/compass.sh add story \
     --slug <slug> \
     --id <capture-uuid> \
     --parent-dir doc/agile/versions/v0/sprint_NN \
     --title "<title>" --description "<description>"
   

   b. Move the capture file into the sprint folder (or delete the original after scaffolding if the ID was preserved):

   git rm doc/agile/product_backlog/next/<slug>.org
   

   c. Fill in the story's * Goal, * Acceptance, and * Tasks sections. At minimum scaffold one task.

7. Update the sprint's * Stories table with each newly added story row (State: BACKLOG, Start: today).

8. Commit all changes:

   git add doc/agile/versions/v0/sprint_NN/ \
           doc/agile/product_backlog/next/ \
           doc/agile/versions/v0/sprint_NN/sprint.org
   git commit -m "[agile] Sprint NN planning — promote N stories from backlog"
   

• How do I plan a sprint? — the human-readable checklist for this skill's procedure.
• How do I create a new doc? — codegen invocations for story scaffolding.
• How do I start work on a story? — once planning is done, this recipe covers BACKLOG → STARTED transition.

• Agile — top of the operational tree.
• Agile process — planning phase owns this skill.
• Product backlog — the next/deferred structure.
• Capture — document type being promoted.
• Story — document type being created.
• Document types — full frontmatter contracts.
• Agile product owner — atomic operations this skill orchestrates.
• Backlog refiner — use instead for general housekeeping without a sprint-selection goal.