Get the FREE Ultimate OpenClaw Setup Guide →

obra-writing-plans

Scanned
npx machina-cli add skill faulkdev/github-copilot-superpowers/obra-writing-plans --openclaw
Files (1)
SKILL.md
3.5 KB

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.

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 obra-brainstorming).

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

Copilot-Friendly Rules

  • Output format: Intent → (optional) Assumptions → numbered Plan → Validation commands.
  • Keep steps concrete and testable; avoid vague verbs.
  • Avoid platform-specific orchestration or slash commands. If parallelism helps, describe it as optional.

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 Copilot:** REQUIRED SKILL: Use `obra-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"

## 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

## Execution Handoff

After saving the plan, offer execution choice:

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

**1. Same session (if available)** - Use `obra-subagent-driven-development` for per-task execution and reviews.

**2. New session** - Open a new session with `obra-executing-plans`, batch execution with checkpoints.

**Which approach?"**

**If Same Session chosen:**
- **REQUIRED SKILL:** Use `obra-subagent-driven-development`
- Stay in this session
- Per-task execution + review

**If New Session chosen:**
- Guide them to open a new session in the worktree
- **REQUIRED SKILL:** New session uses `obra-executing-plans`

Source

git clone https://github.com/faulkdev/github-copilot-superpowers/blob/integrate-obra-superpowers/scripts/overrides/obra/obra-writing-plans/SKILL.mdView on GitHub

Overview

Writing Plans generates comprehensive implementation roadmaps for multi-step tasks in VS Code Copilot Agent mode. It assumes the engineer has zero context, documents every required detail (files, code, tests, docs), and structures work as bite-sized tasks with frequent commits. Plans are saved to docs/plans/YYYY-MM-DD-<feature-name>.md for traceability.

How This Skill Works

When a user requests a plan, the skill outputs a Markdown document with the mandated header and the format: Intent → (optional) Assumptions → numbered Plan → Validation commands. It lists exact files to create or modify, concrete code blocks or snippets, test and doc references, and precise commands. Each task is designed as a 2-5 minute action, and the plan concludes with commit steps and a handoff section for execution in the specified work environment.

When to Use It

  • The user asks for a plan, roadmap, or step-by-step breakdown for a multi-step task in VS Code Copilot Agent mode
  • You need to implement a feature or bug fix that touches multiple files and tests and requires a coordinated plan
  • The engineer has zero context about the codebase and you must document every touchpoint (files, tests, docs) and rationale
  • You want a repeatable, executable plan with concrete file paths, commands, and validation steps
  • You plan to save the document to docs/plans with a date stamp for traceability and auditing

Quick Start

  1. Step 1: Announce use of writing-plans and define the feature name for the plan document
  2. Step 2: List exact files to touch and draft the plan header, then enumerate 5 bite-sized tasks with 2-5 minute scope
  3. Step 3: Save the plan to docs/plans/YYYY-MM-DD-<feature-name>.md and provide the path back to the user

Best Practices

  • Begin with the required plan header and the announce line: I'm using the writing-plans skill to create the implementation plan
  • Use exact file paths and line ranges for all touched files
  • Decompose work into bite-sized steps (2-5 minutes each) and include failing/passing test cycles
  • Include tests, minimal code changes, and explicit commit commands (with messages)
  • DRY, YAGNI, and TDD guidance should guide the plan; avoid unnecessary work and over-engineering

Example Use Cases

  • Plan to add a logging feature to a Python tool, touching src/logging.py, tests/test_logging.py, and docs/logging.md, with explicit run commands
  • Plan to implement a multi-file 'export to CSV' feature in a Node.js service, including service, CLI, tests, and docs
  • Plan to refactor a data pipeline step into smaller modules with unit tests and integration tests
  • Plan to implement a new VS Code Copilot Agent integration in an extension, covering commands, UI, and tests
  • Plan to introduce feature flags and basic telemetry in a React app, with tests and docs

Frequently Asked Questions

Add this skill to your agents
Sponsor this space

Reach thousands of developers