---
name: nanlabs-planning
description: >-
  WHAT - Planning, estimation, task breakdown, and iteration capacity fallback based on NaNLABS Best Practices. Use before finalizing backlog scope, story/task estimates, or iteration commitments.
---

# Planning and Estimation (WHAT)

Use this skill when work needs planning, breakdown, estimation, or capacity validation before it becomes final backlog content.

## Default guardrails

1. Apply **`nanlabs-output-handshake`** before producing any final planning artifact: ask where it should live and confirm human review.
2. Prefer project-specific planning rules if the repo, ticket system, or engagement pack defines them. Otherwise use this default.
3. If a work item is being created or refined, pair this with **`nanlabs-work-item`** and the relevant atomic skill.

## What to check

- **Blast radius** and rollback are explicit for each planned step (see `nanlabs-planner` agent `references/CONTRACT.md`).
- **User approval** before implementation when the plan is non-trivial.
- Scope and goal are clear.
- Work is broken down into manageable items; anything estimated at 5+ points should be considered for splitting.
- Acceptance criteria and dependencies are known enough for estimation.
- Estimation technique is explicit: planning poker, T-shirt sizing, Fibonacci, or the team override.
- Capacity considers availability, holidays, meetings, focus factor, dependencies, and historical velocity.

## References

- `references/default-template.md` - default planning and capacity template
- `references/example-sprint-planning.md` — example sprint plan with capacity math and risk flags
- `nanlabs-work-item` - work item hierarchy and creation
- `clickup-cli` - task updates after user approval