| name | rein-plan |
| description | Use after triage, before implementation. Breaks complex work into sequenced steps with risks, checkpoints, rollback boundaries, and interview-result handoff support. |
| argument-hint | [--from-interview <slug|path>] <task or implementation target> |
rein-plan
Use this after triage and before implementation when the work involves multiple files, multiple concerns, or a non-obvious execution order.
When To Use
- The task touches more than two files or modules
- The execution order matters (migrations before code, types before consumers)
- There are dependencies between subtasks
- The risk of a partial implementation is high
- A completed
rein-interview spec bundle already exists and should drive the plan
When Not To Use
- The task is a single-file, single-concern change
- Triage already produced a clear, linear path
- The user explicitly asks to skip planning
Interview Handoff
Before planning, check whether there is a completed rein-interview artifact to consume.
Artifact priority:
- If the user provides
--from-interview <slug|path>, use that artifact.
- If the user provides or references a
rein interview handoff --to plan result, use its sourceResult and recommendedSkillInvocation.
- Otherwise, look under
.rein/specs/ for rein-interview-*/result.json.
- If one clearly matches the current task, use it.
- If several plausible artifacts exist, prefer the most recent completed one whose
intent, desiredOutcome, or inScope matches the task.
- If no relevant artifact exists, proceed with normal planning.
When a result.json artifact is available:
- treat it as the primary source of truth for intent, desired outcome, in-scope, out-of-scope, constraints, acceptance criteria, assumptions, technical context, and execution bridge
- use
spec.md or the transcript only if result.json is insufficient
- do not silently drift away from the artifact without naming the conflict
- if the live repo facts conflict with the artifact, call out the mismatch in the plan
- if the artifact is missing core planning fields, stop and report that the interview bundle is incomplete rather than planning from a hollow result
Steps
- Load the relevant
rein-interview result.json first when available.
- List every discrete subtask required to complete the work.
- Identify dependencies between subtasks. Order them so each step has its prerequisites satisfied.
- Write each step in
change -> verify form.
- For each step, state:
- what changes
- why this is the minimum sufficient change
- what must stay untouched
- what could go wrong
- how to verify the step succeeded before moving on
- Reject speculative abstraction, flexibility, or helper extraction unless the plan explicitly justifies it.
- Identify the first safe checkpoint — the earliest point where the repo is in a valid state and work can be paused or reviewed.
- Identify rollback boundaries — points where a partial revert is clean rather than destructive.
- Flag any step that requires a decision the user has not yet made.
Output
Emit a plan with:
- ordered step list with per-step minimum-sufficient change, scope boundary, risk, and verification
- dependency graph (if non-linear)
- checkpoints
- rollback boundaries
- open decisions requiring user input
- estimated blast radius (files touched, APIs changed, tests affected)
- when an interview artifact was used, a short note naming the
result.json input and the key handoff fields consumed
Task: {{ARGUMENTS}}