plan-writing
Scannednpx machina-cli add skill a5c-ai/babysitter/plan-writing --openclawPlan Writing
Overview
Convert research findings into actionable implementation plans. Scales planning rigor to stakes level. Every code-changing task specifies tests before implementation.
When to Use
- After research phase identifies what needs to change
- Before implementing any medium or high stakes changes
- When requirements are clear and codebase is understood
Process
- Load research - Find
*-<topic>-research.mdindocs/plans/ - Classify stakes - Low (isolated, reversible), Medium (multiple files), High (architectural)
- Define success criteria - Functional, non-functional, and acceptance criteria
- Decompose tasks - Granular steps with file paths, line references, verification methods
- Plan tests - Test specification as first sub-step per task (test-first)
- Assess risks - Breaking changes, performance, security, dependencies, rollback strategy
- Write plan document -
docs/plans/YYYY-MM-DD-<topic>-plan.md - Approval gate - Human approves, requests changes, or returns to research
Anti-Patterns to Avoid
- Vague task descriptions without specific file references
- Missing verification criteria for any step
- Combining test writing and implementation into single steps
- Planning rigor mismatched to stakes level
- Proceeding without explicit user approval
Tool Use
Invoke via babysitter process: methodologies/rpikit/rpikit-plan
Source
git clone https://github.com/a5c-ai/babysitter/blob/main/plugins/babysitter/skills/babysit/process/methodologies/rpikit/skills/plan-writing/SKILL.mdView on GitHub Overview
Plan-writing translates research findings into concrete, executable plans with stakes-based rigor. It ensures each code-changing task includes tests before implementation and decomposes work into granular steps for clear accountability.
How This Skill Works
It loads research from docs/plans, classifies each change by stake level, defines comprehensive success criteria, and decomposes tasks with precise file paths, line references, and verification methods. It enforces a test-first approach for every task, assesses risks, writes a formal plan to docs/plans/YYYY-MM-DD-<topic>-plan.md, and passes the plan through an approval gate before coding begins.
When to Use It
- After research phase identifies what needs to change
- Before implementing any medium or high-stakes changes
- When requirements are clear and codebase is understood
- When you need explicit success criteria and test-first plan
- When an approval gate is required before coding
Quick Start
- Step 1: Load research from docs/plans/*-research.md to scope changes
- Step 2: Classify stakes, define success criteria, and decompose into granular tasks with file references
- Step 3: Run the rpikit-plan tool to generate docs/plans/YYYY-MM-DD-<topic>-plan.md and submit for approval
Best Practices
- Load the relevant research file from docs/plans to anchor the plan
- Classify stakes accurately (Low, Medium, High) to determine rigor
- Define functional, non-functional, and acceptance criteria before tasks
- Decompose tasks with concrete file references, paths, and verifications
- Plan tests as the first sub-step for every task and secure approval before implementation
Example Use Cases
- Plan to add a small UI tweak with low stakes, including test notes and file references
- Plan to refactor a module across multiple files with clear success criteria
- Plan to add a new API endpoint with test-first tasks and acceptance criteria
- Plan to migrate a database schema with rollback strategy for high-stakes changes
- Plan to overhaul authentication flow with architecture-level changes and risk assessment
