By labrinyang
Brainstorm → Plan → Execute: structured 3-phase development workflow with intelligent execution strategy selection
Implement a specific task following a plan. Use when a well-defined task with clear file paths and acceptance criteria needs to be built. Has write access to create and modify files.
Research a specific topic, investigate codebase patterns, or explore unfamiliar code. Use PROACTIVELY when investigation is needed before implementation. Read-only — does not modify files.
Review code changes against a plan or spec. Use PROACTIVELY after implementation tasks complete. Checks for correctness, edge cases, security, and adherence to the original plan. Read-only.
Use BEFORE writing any code or plan when the user describes a feature, idea, bug, or change request. Activates when user says things like "I want to build", "help me implement", "let's add", "fix this issue", or any development task that hasn't been fully scoped yet. Also use when user says "brainstorm" or uses /bpe:brainstorm.
Use AFTER a plan with task list exists and the user has approved execution. Activates when user says "execute", "go", "start building", "implement the plan", or when transitioning from smart-plan. Also use when user says /bpe:execute.
Use AFTER requirements are validated (design document exists) and BEFORE any implementation begins. Activates when user says "plan", "write a plan", "break this down", or when transitioning from brainstorm-deep. Also use when user says /bpe:plan.
Uses power tools
Uses Bash, Write, or Edit tools
No model invocation
Executes directly as bash, bypassing the AI model
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A Claude Code plugin that enforces structured 3-phase development:
# Step 1: Add marketplace
/plugin marketplace add labrinyang/bpe-method
# Step 2: Install plugin
/plugin install bpe@bpe-marketplace
Just describe what you want to build. BPE skills trigger automatically.
/bpe:brainstorm — Start requirements discovery/bpe:plan — Create implementation plan/bpe:execute — Execute the planEach phase requires explicit user approval to proceed:
brainstorm-deep — Requirements engineeringsmart-plan — Task decomposition and strategy selectionsmart-execute — Implementation coordinationresearcher — Codebase investigation (read-only)builder — Task implementation (read-write)reviewer — Code review (read-only)/bpe:brainstorm, /bpe:plan, /bpe:executeBPE is designed to complement the Superpowers plugin. Install both:
superpowers:test-driven-developmentBPE uses Claude Code's native memory system:
npx claudepluginhub labrinyang/bpe-method --plugin bpeClinical-grade Apple Health analysis: 21 statistical methods, 30 disease screenings, 59 papers cited, 35+ SVG charts, EN/ZH support.
Connect PM, Dev, and QA context in AI-native development. Shared Jira, Confluence, and Figma context across Claude Code sessions — for teams and solo developers working in multiple terminals.
AI agent team orchestration plugin: intelligently routes tasks to subagents or agent teams, composes teams from a role pool based on management theory, and self-evolves through session reflection
Plan and autonomously build a software task end-to-end. Recons the codebase, applies preloaded memory, decomposes into the right number of phases, gets one confirmation, then prepares a single ready-to-paste /goal command — one paste between you and done — that drives execution to completion with built-in retry, fix-spec recovery, and per-phase memory writeback. Works on Claude Code and Codex.
Plan-Build-Run — Structured development workflow for Claude Code. Solves context rot through disciplined subagent delegation, structured planning, atomic execution, and goal-backward verification.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Complete developer toolkit for Claude Code