# Implementation Task Creator

> Use when breaking down a component for implementation - creates task file in implementation_process/in_progress/ with TDD steps and acceptance criteria

- **Type:** Skill
- **Install:** `agentstack add skill-camoa-claude-skills-implementation-task-creator`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [camoa](https://agentstack.voostack.com/s/camoa)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [camoa](https://github.com/camoa)
- **Source:** https://github.com/camoa/claude-skills/tree/main/ai-dev-assistant/skills/implementation-task-creator

## Install

```sh
agentstack add skill-camoa-claude-skills-implementation-task-creator
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Implementation Task Creator

Create task files that break down components into implementable TDD steps.

## Activation

Activate when you detect:
- "Break down X for implementation"
- "Create task for X"
- Moving from Phase 2 to Phase 3
- Need implementation tasks for a component

## Workflow

### 1. Load Component Architecture

Use `Read` on `{project_path}/architecture/{component}.md`

If file doesn't exist, ask: "No architecture found for {component}. Create architecture first?"

Extract:
- Component purpose
- Interface (methods, parameters)
- Dependencies
- Pattern reference
- Acceptance criteria

### 2. Determine Task Scope

Ask user:
```
How should this component be broken down?
1. Single task (small component)
2. Multiple tasks by method/feature
3. Multiple tasks by layer (test, implementation, integration)

Your choice:
```

### 3. Define Task Order

Based on dependencies, determine implementation order. Ask:
```
Proposed task order:
1. {task 1} - {why first}
2. {task 2} - {depends on 1}
3. {task 3} - {depends on 2}

Adjust order? (yes/no)
```

### 4. Create Task File(s)

For each task, use `Write` to create `{project_path}/implementation_process/in_progress/{nn}_{task_name}.md`:

```markdown
# Task: {Task Name}

**Component:** {component name}
**Created:** {YYYY-MM-DD}
**Status:** Not Started
**Priority:** {nn}

## Objective
{What this task accomplishes - one paragraph}

## Prerequisites
- [ ] {Task that must be complete first, or "None"}

## Acceptance Criteria
- [ ] {Criterion 1 - specific and testable}
- [ ] {Criterion 2}
- [ ] {Criterion 3}
- [ ] All tests pass
- [ ] Code follows framework standards

## TDD Steps

### Step 1: Write Failing Test
Create test file:
```
tests/{unit}/{TestClass}Test.{ext}
```

Test case:
```
function test{MethodName}() {
  // Arrange: {setup}
  // Act: {action}
  // Assert: {expected result}
}
```

### Step 2: Verify Test Fails
Run: `ddev {test-runner} {test_path}`
Expected: Test fails (class/method not found)

### Step 3: Write Minimum Implementation
Create: `src/{Type}/{ClassName}.{ext}`
Implement only enough to pass the test.

### Step 4: Verify Test Passes
Run: `ddev {test-runner} {test_path}`
Expected: Test passes

### Step 5: Refactor
Clean up code while keeping tests green.

## Files to Create/Modify
| File | Action | Purpose |
|------|--------|---------|
| `src/{path}` | Create | Main implementation |
| `tests/{path}` | Create | Test coverage |
| `{service-config-file}` | Modify | Service registration |

## Pattern Reference
Follow: `{core path from architecture}`

Key aspects:
- {what to copy from pattern}
- {what to adapt}

## Notes
{Any specific considerations from architecture}
```

### 5. Update project_state.md

Use `Edit` to update the project state:
```markdown
## Current Focus
Implementation Phase - {component name}

## Next Steps
1. Complete task: {first task name}
```

### 6. Confirm

Show user:
```
Created {count} task(s) for {component}:
1. {task 1} - in_progress/{filename}
2. {task 2} - in_progress/{filename}

Start with task 1? Use: /ai-dev-assistant:implement {task_name}
```

## Stop Points

STOP and wait for user:
- If architecture file not found
- After asking about task breakdown
- After showing proposed order
- After creating files (confirm before proceeding)

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [camoa](https://github.com/camoa)
- **Source:** [camoa/claude-skills](https://github.com/camoa/claude-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-camoa-claude-skills-implementation-task-creator
- Seller: https://agentstack.voostack.com/s/camoa
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.