How do I install the Mk Watzup skill?

Run npx skillmds@latest add phuc-nt/mk-watzup in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Mk Watzup safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Mk Watzup?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Mk Watzup free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Mk Watzup?

phuc-nt (@phuc-nt) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Mk Watzup by phuc-nt — Agent Skill

Name: Mk Watzup
Author: phuc-nt

Mk Watzup

Generate short handoff reports from Git branches, remote refs, worktrees, unfinished plans, and roadmap docs. Surfaces priority-ranked next steps with checkbox progress and rationale. Use when the user asks what's in flight, wants progress/next steps, is in a fresh worktree or detached checkout, or needs end-of-session status.

phuc-nt d1bb6f3 6 files · 48.8 KB Updated Sep 17, 2026

File contents

Wrap Up

Create a short, evidence-backed handoff report for the active project, with priority-ranked next steps grounded in plan progress and roadmap state.

This skill handles status and handoff reporting only. It does not implement, edit, commit, checkout, merge, push, or fetch unless the user explicitly requests fresh remote refs.

Required Scan

Run the scanner first from the project root:

node .claude/skills/watzup/scripts/watzup-scan.cjs --json

Use --fetch only when the user asks to refresh remotes before the report:

node .claude/skills/watzup/scripts/watzup-scan.cjs --json --fetch

When developing from this source repository before install, use node claude/skills/watzup/scripts/watzup-scan.cjs --json only if .claude/skills/... is not present.

Default behavior:

Scan local branches and remote branch refs.
Scan registered worktrees.
Scan unfinished plans from visible worktrees and tracked branch refs.
Count - [ ] / - [x] checkboxes in each plan directory (plan.md + phase-*.md) for progress %.
Scan docs/*roadmap*.md and docs/*milestones*.md for active milestones.
Build priority-ranked next steps via composite scoring (see below).
Do not run network operations.
Do not change branches or mutate the checkout.

Priority Ranking

The scanner now emits nextSteps[] as objects with {priority, action, rationale, planId?}. Ordering reflects a composite score per plan:

Status: in-progress (+400) > in-review (+300) > pending (+150).
Workspace alignment: current worktree (+600), current branch (+400).
Provenance: filesystem source (+80), local ref (+40).
Momentum: plans between 40-90% complete get bumped (close to done). Brand-new plans (<10%) get a small starter bump.

Hygiene steps (dirty working tree, detached HEAD) always rank first. Roadmap milestones fill remaining slots after plan-driven actions.

Report Format

Keep output brief. Prefer this structure:

Current State - branch or detached HEAD, dirty/clean, active worktree.
Recent Work - only the highest-signal branches/worktrees.
In-Flight Plans - unfinished plans with X/Y todos · NN% done annotation.
Roadmaps - active milestones from docs/*roadmap*.md, if any.
Next Steps - 5 to 6 priority-ranked actions, each with one-line rationale.
Warnings - scanner failures, stale remote-ref caveat, detached HEAD.

If the scanner fails, say it failed and include the error. Then use minimal read-only fallback commands:

git status --short --branch
git worktree list --porcelain
git for-each-ref --format='%(refname:short) %(committerdate:iso8601) %(objectname:short) %(subject)' refs/heads refs/remotes
find plans -maxdepth 2 -name plan.md -print
find docs -maxdepth 2 -iname '*roadmap*.md' -print

Do not pretend the full scan completed when fallback was used.

---name: mk-watzupdescription: Generate short handoff reports from Git branches, remote refs, worktrees, unfinished plans, and roadmap docs. Surfaces priority-ranked next steps with checkbox progress and rationale. Use when the user asks what's in flight, wants progress/next steps, is in a fresh worktree or detached checkout, or needs end-of-session status.---# Wrap UpCreate a short, evidence-backed handoff report for the active project, with priority-ranked next steps grounded in plan progress and roadmap state.This skill handles status and handoff reporting only. It does not implement, edit, commit, checkout, merge, push, or fetch unless the user explicitly requests fresh remote refs.## Required ScanRun the scanner first from the project root:```bashnode .claude/skills/watzup/scripts/watzup-scan.cjs --json```Use `--fetch` only when the user asks to refresh remotes before the report:```bashnode .claude/skills/watzup/scripts/watzup-scan.cjs --json --fetch```When developing from this source repository before install, use `node claude/skills/watzup/scripts/watzup-scan.cjs --json` only if `.claude/skills/...` is not present.Default behavior:- Scan local branches and remote branch refs.- Scan registered worktrees.- Scan unfinished plans from visible worktrees and tracked branch refs.- Count `- [ ]` / `- [x]` checkboxes in each plan directory (plan.md + phase-*.md) for progress %.- Scan `docs/*roadmap*.md` and `docs/*milestones*.md` for active milestones.- Build priority-ranked next steps via composite scoring (see below).- Do not run network operations.- Do not change branches or mutate the checkout.## Priority RankingThe scanner now emits `nextSteps[]` as objects with `{priority, action, rationale, planId?}`. Ordering reflects a composite score per plan:- **Status**: `in-progress` (+400) > `in-review` (+300) > `pending` (+150).- **Workspace alignment**: current worktree (+600), current branch (+400).- **Provenance**: filesystem source (+80), local ref (+40).- **Momentum**: plans between 40-90% complete get bumped (close to done). Brand-new plans (<10%) get a small starter bump.Hygiene steps (dirty working tree, detached HEAD) always rank first. Roadmap milestones fill remaining slots after plan-driven actions.## Report FormatKeep output brief. Prefer this structure:1. **Current State** - branch or detached HEAD, dirty/clean, active worktree.2. **Recent Work** - only the highest-signal branches/worktrees.3. **In-Flight Plans** - unfinished plans with `X/Y todos · NN% done` annotation.4. **Roadmaps** - active milestones from `docs/*roadmap*.md`, if any.5. **Next Steps** - 5 to 6 priority-ranked actions, each with one-line rationale.6. **Warnings** - scanner failures, stale remote-ref caveat, detached HEAD.If the scanner fails, say it failed and include the error. Then use minimal read-only fallback commands:```bashgit status --short --branchgit worktree list --porcelaingit for-each-ref --format='%(refname:short) %(committerdate:iso8601) %(objectname:short) %(subject)' refs/heads refs/remotesfind plans -maxdepth 2 -name plan.md -printfind docs -maxdepth 2 -iname '*roadmap*.md' -print```Do not pretend the full scan completed when fallback was used.

Frequently asked questions