skills/ralplan/SKILL.md
Consensus planning with adversarial review — planner, architect, and critic iterate until approved
npx skillsauth add sartoris-digital/pi-superpowers ralplanInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Iterative planning workflow where a planner creates, an architect reviews for feasibility, and a critic challenges for quality. The loop continues until the critic approves.
Dispatch planner (standard tier) to create the implementation plan:
subagent({
agent: "planner",
task: "Create an implementation plan for:\n\n[requirements]\n\nContext:\n[codebase context from scout]\n\nReturn a structured plan with numbered tasks, file paths, and verification steps.",
tier: "standard"
})
Dispatch architect (reasoning tier) to review feasibility:
subagent({
agent: "architect",
task: "MODE: plan-review\n\nReview this implementation plan:\n\n[plan from Step 1]\n\nEvaluate feasibility, completeness, ordering, and risks.",
tier: "reasoning"
})
If architect returns concerns or rejected → revise plan and re-submit.
Dispatch critic (reasoning tier) to challenge quality:
subagent({
agent: "critic",
task: "Review this implementation plan:\n\n[plan from Step 1]\n\nArchitect assessment:\n[architect output from Step 2]\n\nProvide adversarial review. You MUST present an alternative approach before approving.",
tier: "reasoning"
})
If critic returns "revise" → incorporate feedback, return to Step 1.
Produce final plan with:
Save to docs/plans/ and offer to execute.
testing
Use when creating new skills, editing existing skills, or verifying skills work before deployment
development
Use when you have a spec or requirements for a multi-step task, before touching code
data-ai
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
tools
Use when starting any conversation - establishes how to find and use skills, requiring Skill tool invocation before ANY response including clarifying questions