By shinpr
Detects shortcut-taking behavior and guides Claude to follow procedures step by step
npx claudepluginhub shinpr/claude-code-workflows --plugin metronome
A Claude Code plugin that prevents Claude from taking shortcuts on repetitive tasks.
When Claude encounters repetitive tasks, it tends to declare "I'll work efficiently" and then:
sed one-liners)git checkout to revert—losing unrelated uncommitted changes tooExample: Asked to fix comments across 20 files one by one, Claude processes 5 files correctly, then announces "I'll handle the rest efficiently." It runs a sed command that corrupts files, notices the breakage, and runs git checkout to undo everything—including your other uncommitted changes.
This plugin detects such declarations and blocks the Bash tool call, reminding Claude to slow down.
| Language | Pattern | Matches |
|---|---|---|
| English | efficien | efficient, efficiently, efficiency |
| 日本語 | 効率 | 効率的, 効率化 |
| 中文 | 高效, 效率 | (2 patterns required) |
| Deutsch | effizien | effizient, Effizienz |
| Français | efficac | efficace, efficacement, efficacité |
| Español / Português | eficien | eficiente, eficientemente, eficiencia |
| 한국어 | 효율 | 효율적으로, 효율화 |
| Русский | эффектив | эффективно, эффективность |
When triggered, Claude receives:
Slow down.
Read the current task, execute it, verify the result, then move to the next.
No prohibitions. Only positive instructions.
Requires Claude Code (this is a Claude Code plugin)
# 1. Start Claude Code
claude
# 2. Install the marketplace
/plugin marketplace add shinpr/metronome
# 3. Install plugin
/plugin install metronome@metronome
# 4. Restart session (required)
# Exit and restart Claude Code
metronome/
├── .claude-plugin/
│ ├── marketplace.json
│ └── plugin.json
├── hooks/
│ └── hooks.json
├── scripts/
│ └── check-efficiency.py
├── tests/
│ └── test_check_efficiency.py
└── README.md
MIT
Enforce mandatory pre-action verification checkpoints to prevent pattern-matching from overriding explicit reasoning. Use this skill when about to execute implementation actions (Bash, Write, Edit) to verify hypothesis-action alignment. Blocks execution when hypothesis unverified or action targets different system than hypothesis identified. Critical for preventing cognitive dissonance where correct diagnosis leads to wrong implementation.
YES.md — PUA says NO, YES says YES. 6-layer AI governance: format → trigger → hooks → anti-slack → gates → memory. Makes AI do things RIGHT with encouragement, not pressure. Available in English, 中文, 日本語.
Operational guidelines: terse responses, sequential execution, no destructive ops without confirmation
Self-evolving Claude Code system that learns from corrections, manages context, and improves every session
Ultra-compressed communication mode. Cuts ~75% of tokens while keeping full technical accuracy by speaking like a caveman.
Executes bash commands
Hook triggers when Bash tool is used
Share bugs, ideas, or general feedback.
Memory compression system for Claude Code - persist context across sessions
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claim