nhouseholder/writing-plans

Writing Plans

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.

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

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

• (User preferences for plan location override this default)

Rigor Level (merged from type-aware-rigor)

Before writing the plan, classify the work type and match depth accordingly:

| Work Type | Plan Depth | QA Tier | Deploy Gate | |-----------|-----------|---------|-------------| | Site update | Bullet list, skip alternatives | Tier 1 (spot-check) | Visual verify | | New feature | Step-by-step with file paths | Tier 2 (functional + edge case) | Baseline + verify | | New build | Detailed with alternatives + risks | Tier 3 (comprehensive) | Full smoke test | | Algorithm/model | Detailed + hypothesis + data validation | Tier 3 + backtest | Backtest + verify | | Campaign/one-off | Minimal — speed over thoroughness | Tier 1 (it works) | Run once, confirm |

Match effort to cost of failure. When ambiguous, default to New feature (Standard).

Scope Check

If the spec covers multiple independent subsystems, it should have been broken into sub-project specs during brainstorming. If it wasn't, suggest breaking this into separate plans — one per subsystem. Each plan should produce working, testable software on its own.

File Structure

Before defining tasks, map out which files will be created or modified and what each one is responsible for. This is where decomposition decisions get locked in.

• Design units with clear boundaries and well-defined interfaces. Each file should have one clear responsibility.
• You reason best about code you can hold in context at once, and your edits are more reliable when files are focused. Prefer smaller, focused files over large ones that do too much.
• Files that change together should live together. Split by responsibility, not by technical layer.
• In existing codebases, follow established patterns. If the codebase uses large files, don't unilaterally restructure - but if a file you're modifying has grown unwieldy, including a split in the plan is reasonable.

This structure informs the task decomposition. Each task should produce self-contained changes that make sense independently.

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 agentic workers:** REQUIRED: Use superpowers:dispatching-parallel-agents (if subagents available) or superpowers:executing-plans to implement this plan. Steps use checkbox (`- [ ]`) syntax for tracking.

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

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

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

---

### Task 0: Switch to Sonnet

- [ ] **STOP. Do not execute this plan on Opus.** Switch to Sonnet now before proceeding. **CLI:** `/model sonnet` · **Desktop/Web app:** select Sonnet from the model picker dropdown. This plan was written by Opus with all decisions pre-made — execution is mechanical. Proceeding on Opus wastes ~40-60% tokens. **Do not continue until model is switched.**

---

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**

```python
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**

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

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

Plan Review Loop

After completing each chunk of the plan:

1. Use the Agent tool to dispatch a plan review subagent. Prompt it with: "Review this implementation plan for completeness, correct ordering, missing edge cases, and realistic scope. Return concerns as a numbered list." Provide the chunk content and path to spec document. If subagents are unavailable, self-review the plan against the original requirements before proceeding.
2. If ❌ Issues Found:
   • Fix the issues in the chunk
   • Re-dispatch reviewer for that chunk
   • Repeat until ✅ Approved
3. If ✅ Approved: proceed to next chunk (or execution handoff if last chunk)

Chunk boundaries: Use ## Chunk N: <name> headings to delimit chunks. Each chunk should be ≤1000 lines and logically self-contained.

Review loop guidance:

• Same agent that wrote the plan fixes it (preserves context)
• If loop exceeds 5 iterations, surface to human for guidance
• Reviewers are advisory - explain disagreements if you believe feedback is incorrect

Execution Handoff

After saving the plan:

"Plan complete and saved to docs/superpowers/plans/<filename>.md. Switch to Sonnet (/model sonnet) before executing — all decisions are baked in, execution is mechanical."

Execution path depends on harness capabilities:

If harness has subagents (Claude Code, etc.):

• REQUIRED: Use superpowers:dispatching-parallel-agents
• Do NOT offer a choice - subagent-driven is the standard approach
• Fresh subagent per task + two-stage review

If harness does NOT have subagents:

• Execute plan in current session using superpowers:executing-plans
• Batch execution with checkpoints for review