Use PROACTIVELY before starting any stretch of autonomous work expected to take a minute or more, such as a large refactor, a multi-file migration, deep research, generating a long document, or any moment where the user will just be waiting and watching. Suggests a one-line workout, then the work starts immediately. Also use when the user asks for a "vibes check", "superset", "workout break", or "give me an exercise".
Scanned 9/6/2026
Install to Claude Code
npx -y skills add YanisMtcr/wait-lifting --skill wait-lifting --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Wait Lifting?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/yanismtcr-wait-lifting)More formats (shields.io, HTML) on the badges page.
---
name: wait-lifting
description: Use PROACTIVELY before starting any stretch of autonomous work expected to take a minute or more, such as a large refactor, a multi-file migration, deep research, generating a long document, or any moment where the user will just be waiting and watching. Suggests a one-line workout, then the work starts immediately. Also use when the user asks for a "vibes check", "superset", "workout break", or "give me an exercise".
---
# Wait Lifting
Turn the user's wait time into workout time. Before starting a long stretch of autonomous work, tell the user to exercise while they wait, then start the work immediately.
## When to fire
- About to start work estimated at **1 minute or more** with no user input needed: big refactors, multi-file edits, long research, large document generation.
- The user explicitly asks for an exercise, a "vibes check", or a "superset".
## When NOT to fire
- The estimated work is under a minute.
- The hook fired in the last 10 minutes. Its messages are invisible to you, so check the shared cooldown file instead: run `find ~/.claude/wait-lifting.last -mmin -10` and stay silent if it prints the path.
- You are about to launch a subagent (Task/Agent tool): the hook already covers that case, so don't double up.
- The user is mid-conversation and about to answer a question. Never make them exercise while you wait for THEM.
## How to dose
Scale reps to the estimated duration. Pick ONE exercise at random from this table (or the circuit for 5 min+):
| Estimated wait | Push-ups | Squats | Plank | Lunges (each leg) | Burpees |
|----------------|----------|--------|-------|-------------------|---------|
| ~1 min | 5 | 10 | 20s | 6 | 3 |
| ~2 min | 10 | 20 | 40s | 10 | 6 |
| ~3 min | 15 | 30 | 60s | 15 | 9 |
| 5 min + | Mini-circuit: 10 push-ups, 20 squats, 30s plank | | | | |
## Tone
One line, in English, with the estimated duration. Match these examples:
- "This refactor will take me about 3 minutes. 15 push-ups while you wait?"
- "About 2 min of research ahead, just enough for 20 squats."
- "I'll write the doc (~5 min), you do the circuit: 10 push-ups, 20 squats, 30s plank."
## After suggesting
Update the shared state so the hook stays quiet and the status line shows the coach during the wait. Split your message into an info part and a workout part, then run:
```bash
python3 - <<'EOF'
import json, time
from pathlib import Path
d = Path.home() / ".claude"
(d / "wait-lifting.last").write_text(str(time.time()))
(d / "wait-lifting.msg").write_text(json.dumps({
"info": "<info part>",
"workout": "<workout part>",
"expires": time.time() + <estimated minutes> * 60,
}))
EOF
```
## Rules
- One line only, then start the work immediately; never wait for the user to finish exercising.
- Never block, delay, or gate the actual task on the exercise.
- If the user says stop, ignores it, or seems annoyed, drop the suggestions for the rest of the session.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!