71-close-tidy-150
[71] CLOSE. Quick, safe cleanup after completing a milestone. Fix objective issues only (syntax errors, dead code, poor naming). Must be <5% of main task time, <30 seconds per fix, and reversible. Use after key points, not after every small change.
Best use case
71-close-tidy-150 is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
[71] CLOSE. Quick, safe cleanup after completing a milestone. Fix objective issues only (syntax errors, dead code, poor naming). Must be <5% of main task time, <30 seconds per fix, and reversible. Use after key points, not after every small change.
Teams using 71-close-tidy-150 should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/71-close-tidy-150/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How 71-close-tidy-150 Compares
| Feature / Agent | 71-close-tidy-150 | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
[71] CLOSE. Quick, safe cleanup after completing a milestone. Fix objective issues only (syntax errors, dead code, poor naming). Must be <5% of main task time, <30 seconds per fix, and reversible. Use after key points, not after every small change.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Close-Tidy 150 Protocol **Core Principle:** Clean as you go — but safely. After milestones, fix obvious issues quickly. Objective problems only. Time-boxed. Reversible. User-approved. ## What This Skill Does When you invoke this skill, you're asking AI to: - **Scan for obvious issues** — Syntax, dead code, naming - **Time-box strictly** — ≤5% of main task time - **Stay safe** — Only reversible, objective fixes - **Get approval** — User confirms before execution - **Document impact** — Report what was improved ## The 150% Tidy Rule | Dimension | 100% Core | +50% Enhancement | |-----------|-----------|------------------| | **Issues** | Objective defects only | + Prove each is defect | | **Time** | ≤5% of main task | + <30 sec per fix | | **Safety** | All fixes reversible | + Tests still pass | | **Scope** | No expansion | + User approves list | ## What Qualifies as Tidy-Up ``` ✅ ALLOWED (Objective Defects) ├── Syntax Errors: Clear compilation issues ├── Dead Code: Unused, unreachable code ├── Poor Naming: Confusing variable/function names ├── Unused Imports: Import statements not used ├── Obvious Typos: Clear spelling mistakes └── Simple Formatting: Obvious style violations ❌ NOT ALLOWED (Scope Creep) ├── Refactoring: Changing code structure ├── New Features: Adding functionality ├── Optimization: Performance improvements ├── Architecture: Changing design patterns ├── Complex Changes: Anything needing analysis └── Debatable Issues: Subjective improvements ``` ## Time Limits | Fix Type | Time Limit | Safety Check | Revert Ease | |----------|------------|--------------|-------------| | **Syntax Fix** | <10 sec | Auto-check | Instant | | **Naming Fix** | <15 sec | Code review | Instant | | **Dead Code** | <20 sec | Reference check | Instant | | **Unused Import** | <10 sec | Compile check | Instant | | **Simple Format** | <15 sec | Visual check | Instant | | **Complex Change** | ❌ Forbidden | N/A | N/A | **Rule:** If it takes >30 seconds to verify safety → NOT a tidy-up item. ## When to Use This Skill **TRIGGER:** Only after **Key Point Milestones**: - ✅ Feature implementation complete - ✅ Major refactoring done - ✅ Bug fix verified - ✅ Phase of plan completed **NOT TRIGGER:** - ❌ After every file edit - ❌ During active development - ❌ Before understanding the code - ❌ When unsure about impact ## Execution Protocol ### Step 1: MILESTONE CHECK ``` 🏁 **MILESTONE VERIFICATION** **Completed:** [What milestone was reached] **Main Task Time:** [How long the main work took] **Tidy Budget:** [5% of main task = X minutes] ``` ### Step 2: OBSERVATION SCAN Review for obvious issues: ``` 🔍 **SCAN RESULTS** **Issues Found:** 1. [Issue]: [Location] - [Fix time estimate] 2. [Issue]: [Location] - [Fix time estimate] 3. [Issue]: [Location] - [Fix time estimate] **Total Fixes:** [N] **Total Time Estimate:** [X minutes] **Within Budget:** ✅ Yes | ❌ No (reduce scope) ``` ### Step 3: SAFETY VERIFICATION For each issue: ``` 🛡️ **SAFETY CHECK** Issue: [Description] ├── Objective Defect: ✅ Provable | ❌ Subjective ├── Reversible: ✅ Easy revert | ❌ Complex ├── Tests Pass: ✅ Verified | ⚠️ Need to check └── No Side Effects: ✅ Contained | ❌ Cascading Safe to Fix: ✅ Yes | ❌ No ``` ### Step 4: CLEANUP PACKAGE Present for approval: ``` 🧹 **TIDY-UP PROPOSAL** **Milestone:** [What was completed] **Time Budget:** [X minutes] (5% of main task) **Proposed Fixes:** 1. ✅ [Fix 1]: [Description] - [X sec] 2. ✅ [Fix 2]: [Description] - [X sec] 3. ✅ [Fix 3]: [Description] - [X sec] **Total Time:** [Y minutes] **All Reversible:** ✅ Yes **All Objective:** ✅ Yes **Approve cleanup?** (Yes / No / Modify list) ``` ### Step 5: CONTROLLED EXECUTION Apply fixes one by one: - Execute single fix - Verify immediately - Document change - Stop if issues arise ### Step 6: REPORT ``` 🧹 **TIDY-UP 150 COMPLETE** **Fixes Applied:** ✅ [Fix 1]: [What was done] ✅ [Fix 2]: [What was done] ✅ [Fix 3]: [What was done] **Time Spent:** [X minutes] ([Y% of budget]) **Verification:** ├── Tests: ✅ Passing ├── Functionality: ✅ Preserved ├── Revert Ready: ✅ Yes └── No Side Effects: ✅ Confirmed **Impact:** ├── Code Quality: Improved ├── Technical Debt: Reduced └── Future Benefit: [Description] ``` ## Output Format Proposal: ``` 🧹 **TIDY-UP 150 PROPOSAL** **After Milestone:** [What was completed] **Budget:** [X min] (5% of [Y min] main task) **Fixes:** | # | Issue | Location | Time | Safe | |---|-------|----------|------|------| | 1 | [Issue] | [File:Line] | Xs | ✅ | | 2 | [Issue] | [File:Line] | Xs | ✅ | **Total:** [X sec] | **All Safe:** ✅ **Approve?** (Yes / No / Modify) ``` Report: ``` 🧹 **TIDY-UP 150 DONE** **Applied:** [N] fixes in [X] minutes **Tests:** ✅ Passing **Quality:** Improved **Changes:** - [File]: [What changed] - [File]: [What changed] ``` ## Operational Rules 1. **KEY POINT ONLY:** Trigger only after major milestones 2. **TIME BOUND:** Never exceed 5% of main task time 3. **OBJECTIVE ONLY:** Provable defects, not opinions 4. **SAFETY FIRST:** Every fix must be verifiable safe 5. **USER APPROVAL:** Get permission before executing 6. **SCOPE CONTROL:** No expansion beyond identified issues ## Failure Modes & Recovery | Failure | Detection | Recovery | |---------|-----------|----------| | **Scope Creep** | Fixing more than listed | Stop, create separate task | | **Time Overrun** | Exceeding 5% budget | Pause, reschedule remaining | | **Safety Breach** | Fix introduces issues | Immediate revert | | **Unapproved** | Fixing without consent | Revert, get approval | ## Examples ### ❌ Bad Tidy-Up ``` Milestone: Small bug fix (10 minutes) "Tidy-up": - Refactored entire module architecture - Added new helper functions - Changed error handling approach Time: 3 hours (1800% of main task!) Result: Introduced new bugs, delayed delivery ``` ### ✅ Good Tidy-Up ``` 🧹 TIDY-UP 150 PROPOSAL After Milestone: Feature implementation (2 hours) Budget: 6 min (5% of 120 min) Fixes: | # | Issue | Location | Time | Safe | |---|-------|----------|------|------| | 1 | Unused import | auth.ts:3 | 5s | ✅ | | 2 | Typo in var name | user.ts:45 | 10s | ✅ | | 3 | Dead function | utils.ts:89 | 15s | ✅ | Total: 30 sec | All Safe: ✅ User: "Yes" 🧹 TIDY-UP 150 DONE Applied: 3 fixes in 30 seconds Tests: ✅ Passing Quality: Improved Changes: - auth.ts: Removed unused 'lodash' import - user.ts: Renamed 'usrData' → 'userData' - utils.ts: Removed unused 'legacyFormat()' function ``` ## Relationship to Other Skills - **gated-exec-150** → Completes main work - **tidy-up-150** → Quick cleanup after milestone - **integrity-check-150** → Full quality check --- **Remember:** Tidy-up is housekeeping, not renovation. Quick fixes for obvious issues. If you're thinking about it for more than 30 seconds, it's not a tidy-up item — it's a separate task.
Related Skills
72-close-session-150
[72] CLOSE. Save and restore session context between conversations. Use when ending a session to preserve progress, or starting a new session to restore context. Triggers on "save session", "end session", "preserve context", "handoff", "continue from last time", or when context window is running low.
70-close-task-150
[70] CLOSE. Manage task lifecycle with correct statuses and no duplicates. Ensure each task has one status, artifacts match claims, and no task appears in multiple places. Use when managing task lists, updating progress, or ensuring task tracking integrity across systems.
[invalid yaml with unclosed bracket
This YAML is malformed
close-automation
Automate Close CRM tasks via Rube MCP (Composio): create leads, manage calls/SMS, handle tasks, and track notes. Always search tools first for current schemas.
bgo
Automates the complete Blender build-go workflow, from building and packaging your extension/add-on to removing old versions, installing, enabling, and launching Blender for quick testing and iteration.
ask-expert
Creates expert consultation documents with code extraction, git diffs, and size tracking (125KB limit). Use when user asks to "create an expert consultation document", "prepare code for expert review", "gather architecture context", or needs comprehensive technical documentation for external analysis. Requires Node.js 18+.
ask-codex
Asks Codex CLI for coding assistance. Use for getting a second opinion, code generation, debugging, or delegating coding tasks.
ask-claude
Asks Claude CLI for coding assistance. Use for getting a second opinion, code generation, debugging, or delegating coding tasks.
asin-data-api-automation
Automate Asin Data API tasks via Rube MCP (Composio). Always search tools first for current schemas.
ash-vibez
Fetch Ash Framework guidance for creating and managing Ash applications. Use when working with Ash Framework, creating new Ash projects, adding Ash extensions, or needing Ash-specific patterns.
ash-phoenix
AshPhoenix integration guidelines for using Ash Framework with Phoenix. Use when working with AshPhoenix.Form, creating forms backed by Ash resources, handling nested forms, union types in forms, or integrating Ash actions with Phoenix LiveViews. Covers form creation, validation, submission, and error handling patterns.
ash-library-hotfix
Handles emergency hotfix process for critical bugs in ash_cookie_consent library including branch creation, minimal fixes, testing, and rapid release. Use when user asks to "create hotfix", "emergency fix", "patch critical bug", or "hotfix for version".