<!-- TOKEN BUDGET: 220 lines / ~660 tokens -->

Writing Plans

<activation>

When This Skill Activates

• You have a spec, requirements, or design for a multi-step implementation task
• You need to break work into bite-sized, executable tasks before touching code
• You are preparing work for builder agents or a parallel execution session

Announce at start: "I'm using the writing-plans skill to create the implementation plan."

Context: This should be run in a dedicated worktree (created by brainstorming skill).

Save plans to: docs/plans/YYYY-MM-DD-<feature-name>.md

Natural Language Triggers

• "write a plan", "create a plan", "plan this feature", "break this down into tasks"

</activation> <instructions>

Overview

Write comprehensive implementation plans assuming the engineer has zero context for our codebase and questionable taste. Document everything they need to know: which files to touch for each task, code, testing, docs they might need to check, how to test it. Give them the whole plan as bite-sized tasks. DRY. YAGNI. TDD. Frequent commits.

Assume they are a skilled developer, but know almost nothing about our toolset or problem domain. Assume they don't know good test design very well.

Shipyard Plan Format

Shipyard plans use XML-structured tasks with verification criteria. Each task includes:

<task id="1" name="Component Name">
  <description>What this task accomplishes</description>
  <files>
    <create>exact/path/to/file.py</create>
    <modify>exact/path/to/existing.py:123-145</modify>
    <test>tests/exact/path/to/test.py</test>
  </files>
  <steps>
    <step>Write the failing test</step>
    <step>Run test to verify it fails</step>
    <step>Write minimal implementation</step>
    <step>Run test to verify it passes</step>
    <step>Commit</step>
  </steps>
  <verification>
    <command>pytest tests/path/test.py::test_name -v</command>
    <expected>PASS</expected>
  </verification>
</task>

This structured format enables /shipyard:build to parse and execute tasks systematically, and /shipyard:status to track progress.

Bite-Sized Task Granularity

Each step is one action (2-5 minutes):

• "Write the failing test" - step
• "Run it to make sure it fails" - step
• "Implement the minimal code to make the test pass" - step
• "Run the tests and make sure they pass" - step
• "Commit" - step

Plan Document Header

Every plan MUST start with this header:

# [Feature Name] Implementation Plan

> **For Claude:** REQUIRED SUB-SKILL: Use shipyard:shipyard-executing-plans to implement this plan task-by-task.

**Goal:** [One sentence describing what this builds]

**Architecture:** [2-3 sentences about approach]

**Tech Stack:** [Key technologies/libraries]

---

Task Structure

### Task N: [Component Name]

**Files:**
- Create: `exact/path/to/file.py`
- Modify: `exact/path/to/existing.py:123-145`
- Test: `tests/exact/path/to/test.py`

**Step 1: Write the failing test**

```python
def test_specific_behavior():
    result = function(input)
    assert result == expected

Step 2: Run test to verify it fails

Run: pytest tests/path/test.py::test_name -v Expected: FAIL with "function not defined"

Step 3: Write minimal implementation

def function(input):
    return expected

Step 4: Run test to verify it passes

Run: pytest tests/path/test.py::test_name -v Expected: PASS

Step 5: Commit

git add tests/path/test.py src/path/file.py
git commit -m "feat: add specific feature"

</instructions>

<examples>

## Example: Well-Written vs Poorly-Written Plan Task

<example type="good" title="Clear, executable task with exact paths and code">
### Task 2: Add Email Validation

**Files:**
- Create: `src/validators/email.py`
- Test: `tests/validators/test_email.py`

**Step 1: Write the failing test**
```python
def test_rejects_empty_email():
    with pytest.raises(ValidationError, match="email is required"):
        validate_email("")

Step 2: Run test to verify it fails Run: pytest tests/validators/test_email.py::test_rejects_empty_email -v Expected: FAIL with "cannot import name 'validate_email'"

Step 3: Write minimal implementation

from src.errors import ValidationError

def validate_email(email: str) -> str:
    if not email or not email.strip():
        raise ValidationError("email is required")
    return email.strip()

Step 4: Run test to verify it passes Run: pytest tests/validators/test_email.py::test_rejects_empty_email -v Expected: PASS

Step 5: Commit

git add src/validators/email.py tests/validators/test_email.py
git commit -m "feat: add email validation with empty check"

</example> <example type="bad" title="Vague task that leaves builder guessing"> ### Task 2: Add Validation

Add email validation to the project. Make sure it handles edge cases. Write tests. </example>

The good example provides exact file paths, complete code, exact commands with expected output, and a single commit per TDD cycle. The bad example forces the builder to guess paths, invent code, and figure out what "edge cases" means.

</examples> <rules>

Remember

• Exact file paths always
• Complete code in plan (not "add validation")
• Exact commands with expected output
• Reference relevant skills with @ syntax
• DRY, YAGNI, TDD, frequent commits

</rules>

Execution Handoff

After saving the plan, offer execution choice:

"Plan complete and saved to docs/plans/<filename>.md. Two execution options:

1. Agent-Driven (this session) - I dispatch fresh builder agent per task, review between tasks, fast iteration

2. Parallel Session (separate) - Open new session with executing-plans, batch execution with checkpoints

Which approach?"

If Agent-Driven chosen:

• REQUIRED SUB-SKILL: Use shipyard:shipyard-executing-plans
• Stay in this session
• Fresh builder agent per task + two-stage review (spec compliance then code quality)

If Parallel Session chosen:

• Guide them to open new session in worktree
• REQUIRED SUB-SKILL: New session uses shipyard:shipyard-executing-plans