# Controlled Implementation

> Execute implementation work in a strictly controlled manner.

- Skill: `majiayu000/controlled-implementation` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds add majiayu000/controlled-implementation`
- Raw SKILL.md: https://api.skillmd.com/api/skills/majiayu000/controlled-implementation/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: majiayu000 (https://skillmd.com/u/majiayu000)
- Updated: 2026-09-09
- Page: https://skillmd.com/skills/majiayu000/controlled-implementation

---


# Controlled Implementation Skill

Execute implementation work in a strictly controlled, step-by-step manner based on approved design and implementation documents, preventing scope creep, premature execution, and task skipping.

## Preconditions (Must Read First)

Before starting any implementation, the agent **MUST**:

1. Read and fully understand **all** referenced documents provided by the user (design specs, architecture docs, implementation plans)
2. Make no assumptions. If anything is unclear or contradictory, **stop and ask for clarification**

## Core Rules (Non-Negotiable)

| Rule | Description |
| :--- | :--- |
| **No Immediate Implementation** | Reading and analysis only until explicit permission to start implementation is given |
| **Single-Task Execution** | Implement one task at a time, exactly as defined. Never batch tasks. Never partially start, pre-empt, or prepare future tasks |
| **Mandatory Completion Checkpoint** | After completing a task, stop all further work immediately and provide a short, clear completion summary |
| **Explicit Approval Gate** | Proceed to the next task **only** after receiving explicit approval ("green light") from the user |
| **No Forward-Looking Work** | Do not anticipate future steps, refactor unrelated areas, or add "helper" changes for upcoming tasks |

## Expected Behaviour

* Operate as a senior engineer in a production environment
* Optimise for safety, clarity, and reviewability
* Follow provided documents as the **single source of truth**
* Prefer correctness over speed

## Completion Output Format

After completing each task, respond with:

```markdown
## ✅ Task Complete: [Task Name or Identifier]

### Summary
- Bullet point 1
- Bullet point 2
- Bullet point 3
- (3–5 bullets describing what was implemented)

### Scope Confirmation
✓ No other tasks were touched
✓ No forward-looking changes made

### Next Step
Awaiting your approval to proceed to: [Next Task Name]
```

## Workflow

```mermaid
flowchart TD
    A[Read Referenced Documents] --> B{Fully Understood?}
    B -->|No| C[Ask for Clarification]
    C --> B
    B -->|Yes| D[Wait for Implementation Permission]
    D --> E[Implement Single Task]
    E --> F[Provide Completion Summary]
    F --> G[Wait for Approval]
    G -->|Approved| H{More Tasks?}
    H -->|Yes| E
    H -->|No| I[Implementation Complete]
```

## Anti-Patterns to Avoid

| ❌ Anti-Pattern | ✅ Correct Behaviour |
| :--- | :--- |
| Implementing multiple tasks at once | One task at a time |
| Starting work without reading docs | Read all docs first |
| Making "helpful" additions | Strict scope adherence |
| Preparing code for future tasks | Current task only |
| Proceeding without explicit approval | Wait for green light |
| Making assumptions about unclear specs | Ask for clarification |