{"id":45691541,"url":"https://github.com/sh3rd3n/megazord","last_synced_at":"2026-02-24T18:02:12.887Z","repository":{"id":339683539,"uuid":"1161992964","full_name":"Sh3rd3n/megazord","owner":"Sh3rd3n","description":"Claude Code framework for project management, code quality, and multi-agent coordination","archived":false,"fork":false,"pushed_at":"2026-02-20T23:43:09.000Z","size":257,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"master","last_synced_at":"2026-02-21T05:06:23.928Z","etag":null,"topics":["ai-agent","claude-code","claude-code-plugin","cli","code-quality","multi-agent","nodejs","project-management","typescript","workflow-automation"],"latest_commit_sha":null,"homepage":null,"language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Sh3rd3n.png","metadata":{"files":{"readme":"README.md","changelog":"CHANGELOG.md","contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-02-19T18:50:19.000Z","updated_at":"2026-02-20T23:41:58.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/Sh3rd3n/megazord","commit_stats":null,"previous_names":["sh3rd3n/megazord"],"tags_count":2,"template":false,"template_full_name":null,"purl":"pkg:github/Sh3rd3n/megazord","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sh3rd3n%2Fmegazord","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sh3rd3n%2Fmegazord/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sh3rd3n%2Fmegazord/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sh3rd3n%2Fmegazord/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Sh3rd3n","download_url":"https://codeload.github.com/Sh3rd3n/megazord/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Sh3rd3n%2Fmegazord/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29793659,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-24T16:37:37.581Z","status":"ssl_error","status_checked_at":"2026-02-24T16:37:37.074Z","response_time":75,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["ai-agent","claude-code","claude-code-plugin","cli","code-quality","multi-agent","nodejs","project-management","typescript","workflow-automation"],"created_at":"2026-02-24T18:02:08.026Z","updated_at":"2026-02-24T18:02:12.880Z","avatar_url":"https://github.com/Sh3rd3n.png","language":"TypeScript","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://raw.githubusercontent.com/Sh3rd3n/megazord/master/assets/megazord-hero.svg\" alt=\"Megazord\" width=\"820\"\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cstrong\u003eProject management, code quality, and multi-agent coordination in a single Claude Code plugin.\u003c/strong\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"https://www.npmjs.com/package/megazord-cli\"\u003e\u003cimg src=\"https://img.shields.io/npm/v/megazord-cli?style=flat\u0026colorA=18181B\u0026colorB=28CF8D\" alt=\"npm version\"\u003e\u003c/a\u003e\n  \u003ca href=\"https://github.com/sh3rd3n/megazord/actions\"\u003e\u003cimg src=\"https://img.shields.io/github/actions/workflow/status/sh3rd3n/megazord/ci.yml?branch=master\u0026label=CI\u0026style=flat\u0026colorA=18181B\" alt=\"CI\"\u003e\u003c/a\u003e\n  \u003ca href=\"LICENSE\"\u003e\u003cimg src=\"https://img.shields.io/badge/License-MIT-blue.svg?style=flat\u0026colorA=18181B\" alt=\"License: MIT\"\u003e\u003c/a\u003e\n  \u003ca href=\"https://nodejs.org/\"\u003e\u003cimg src=\"https://img.shields.io/badge/Node.js-%3E%3D22-green?style=flat\u0026colorA=18181B\" alt=\"Node.js \u003e= 22\"\u003e\u003c/a\u003e\n  \u003ca href=\"https://claude.ai/code\"\u003e\u003cimg src=\"https://img.shields.io/badge/Claude_Code-Plugin-purple?style=flat\u0026colorA=18181B\" alt=\"Claude Code\"\u003e\u003c/a\u003e\n  \u003cimg src=\"https://img.shields.io/badge/Commands-15-orange?style=flat\u0026colorA=18181B\" alt=\"Commands: 15\"\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003ccode\u003ebunx megazord-cli\u003c/code\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003ca href=\"#why-megazord\"\u003eWhy Megazord\u003c/a\u003e \u0026middot;\n  \u003ca href=\"#how-it-works\"\u003eHow It Works\u003c/a\u003e \u0026middot;\n  \u003ca href=\"#cortex\"\u003eCORTEX\u003c/a\u003e \u0026middot;\n  \u003ca href=\"#commands\"\u003eCommands\u003c/a\u003e \u0026middot;\n  \u003ca href=\"#why-it-works\"\u003eWhy It Works\u003c/a\u003e \u0026middot;\n  \u003ca href=\"#configuration\"\u003eConfiguration\u003c/a\u003e\n\u003c/p\u003e\n\n---\n\n## Why Megazord\n\nOther Claude Code frameworks fall into two camps. **Workflow tools** (like Superpowers) give you code quality discipline -- TDD, brainstorming, code review -- but don't help you manage a project across phases and milestones. **Project management tools** (like GSD) give you roadmaps, planning, and context engineering -- but leave code quality to chance.\n\nMegazord combines both. One plugin that covers the full development lifecycle: from \"what should I build?\" all the way to \"it's built, reviewed, verified, and shipped.\"\n\nAnd unlike other frameworks where agents are spawned, do their job, and disappear -- Megazord is built on Claude Code's **native Agent Teams**. Your agents actually communicate with each other, coordinate through a shared task list, and hand off context in real time. No file-based workarounds. No fire-and-forget subagents. Real multi-agent coordination.\n\n---\n\n## Who This Is For\n\nDevelopers who use Claude Code and want their AI-assisted workflow to feel like a coordinated team instead of a solo assistant. Whether you're building a new project from scratch or adding features to an existing codebase, Megazord gives you structure without bureaucracy.\n\n---\n\n## Getting Started\n\n### Prerequisites\n\n- **[Claude Code](https://claude.ai/code)** \u003e= 2.x\n- **[Node.js](https://nodejs.org/)** \u003e= 22\n- **[bun](https://bun.sh/)** (any recent version)\n\n### Install\n\n```bash\nbunx megazord-cli\n```\n\nThis registers Megazord as a Claude Code plugin. Restart Claude Code to activate.\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eOther install commands\u003c/strong\u003e\u003c/summary\u003e\n\n```bash\n# Update to latest version\nbunx megazord-cli update\n\n# Uninstall\nbunx megazord-cli uninstall\n```\n\n\u003c/details\u003e\n\n### Your first project\n\n**1. Initialize**\n\n```\n/mz:init\n```\n\nMegazord asks about your project, picks a quality preset, and creates the planning structure.\n\n**2. Plan your first phase**\n\n```\n/mz:plan 1\n```\n\nResearches the technical landscape, then decomposes the phase into tasks with dependencies and execution waves.\n\n**3. Build it**\n\n```\n/mz:go\n```\n\nSpawns specialized agents that execute tasks in dependency order, commit atomically, and coordinate through Agent Teams.\n\n**4. Verify it works**\n\n```\n/mz:verify\n```\n\nChecks that what was built actually matches your acceptance criteria -- not just \"did tasks complete\" but \"did we build what we said we'd build.\"\n\n---\n\n## How It Works\n\n### 1. Initialize Project\n\n```\n/mz:init\n```\n\nOne command, one flow. The system:\n\n1. **Questions** -- asks until it understands your project completely (goals, constraints, tech stack, edge cases)\n2. **Preset selection** -- Strict, Balanced, or Minimal quality gates\n3. **Deep context** -- captures your vision, requirements, and non-goals\n4. **Planning structure** -- creates PROJECT.md, STATE.md, ROADMAP.md, and config\n\nYou approve the setup. Now you're ready to build.\n\n**Creates:** `.planning/megazord.config.json`, `PROJECT.md`, `STATE.md`, `ROADMAP.md`\n\n\u003e **Already have code?** Run `/mz:map` first. It spawns parallel agents to analyze your stack, architecture, conventions, and concerns. Then `/mz:init` and `/mz:plan` know your codebase -- planning automatically loads your existing patterns.\n\n---\n\n### 2. Discuss Phase (optional)\n\n```\n/mz:discuss \"Should we use JWT or session-based auth?\"\n```\n\n**This is where you shape the implementation.**\n\nYour roadmap has a sentence or two per phase. That's not enough context to build something the way *you* imagine it. This step captures your preferences before anything gets researched or planned.\n\nMegazord identifies gray areas, asks Socratic questions, explores 5+ approaches, and records your decisions. The output feeds directly into research and planning.\n\n**Creates:** `CONTEXT.md`\n\n---\n\n### 3. Plan Phase\n\n```\n/mz:plan 1\n```\n\nThe system:\n\n1. **Researches** -- investigates how to implement this phase, guided by your decisions\n2. **Plans** -- creates task breakdown with dependencies and execution waves\n3. **Verifies** -- checks plans against requirements, loops until they pass\n\nEach plan is structured so agents can execute independently with clear boundaries.\n\n**Creates:** `RESEARCH.md`, `PLAN.md`\n\n---\n\n### 4. Execute Phase\n\n```\n/mz:go\n```\n\nThe system:\n\n1. **Runs tasks in waves** -- parallel where independent, sequential when dependent\n2. **Spawns specialized agents** -- each task gets an executor with fresh context\n3. **Commits per task** -- every completed task gets its own atomic commit\n4. **Coordinates via Agent Teams** -- agents communicate, hand off context, and build on each other's work\n\nWalk away, come back to completed work with clean git history.\n\n**How wave execution works:**\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://raw.githubusercontent.com/Sh3rd3n/megazord/master/assets/wave-execution.svg\" alt=\"Wave Execution Flow\" width=\"820\"\u003e\n\u003c/p\u003e\n\nIndependent tasks run in parallel within each wave. Waves run sequentially -- Wave 2 waits for Wave 1 to finish.\n\n**Creates:** `SUMMARY.md`, `VERIFICATION.md`\n\n---\n\n### 5. Verify and Review\n\n```\n/mz:verify\n/mz:review\n```\n\n**Verify** checks that deliverables match your acceptance criteria. Goal-backward: \"did we build what we promised?\"\n\n**Review** runs a two-stage code review: spec compliance first, then code quality. Findings are severity-graded -- critical issues block progress.\n\n---\n\n### 6. Repeat, Complete, Next Milestone\n\n```\n/mz:plan 2 → /mz:go → /mz:verify → ...\n/mz:lifecycle\n```\n\nLoop **discuss/plan/execute/verify** until milestone complete. Then `/mz:lifecycle` audits deliverables, archives the milestone, tags the release, and prepares the next version.\n\n---\n\n### Quick Mode\n\n```\n/mz:quick \"Add dark mode toggle to settings page\"\n```\n\nFor tasks that don't need full planning. Same quality guarantees (atomic commits, state tracking), faster path. No research, no plan checker, no verifier -- just get it done.\n\n---\n\n## CORTEX\n\n\u003e *The adaptive reasoning engine that makes agents think before they act.*\n\nMost AI coding tools treat every task the same. A variable rename gets the same cognitive process as a distributed authentication system. That's wasteful at the bottom and dangerous at the top.\n\nCORTEX classifies every task through **Cynefin complexity domains** -- Clear, Complicated, Complex, Chaotic -- and scales analysis to match actual risk. Clear tasks execute instantly with zero overhead. Complicated tasks get structured challenge analysis. Complex tasks trigger full decomposition before a single line of code is written. Chaotic situations stop execution entirely and escalate to the human.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://raw.githubusercontent.com/Sh3rd3n/megazord/master/assets/cortex-diagram.svg\" alt=\"CORTEX Cynefin Domain Flow\" width=\"820\"\u003e\n\u003c/p\u003e\n\n### Thinking Frameworks\n\n| Domain | Framework | What It Prevents |\n|--------|-----------|-----------------|\n| Complicated+ | Inversion (Pre-mortem) | Building something that fails in predictable ways |\n| Complicated+ | Ladder of Inference | Acting on unverified assumptions |\n| Complicated+ | Consequence Tracing | Missing second and third-order effects |\n| Complex | First-Principles Decomposition | Borrowing solutions that don't fit |\n| Complex | Abstraction Laddering | Solving the wrong problem |\n| Complex | Iceberg Model | Treating symptoms instead of structural causes |\n| Recurring | Issue Tree | Debugging without structure |\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eInversion (Pre-mortem)\u003c/strong\u003e\u003c/summary\u003e\n\nBefore writing code, assume the implementation has already failed. Generate three specific failure scenarios:\n\n1. *\"This fails when...\"* -- not vague risks, but concrete failure paths\n2. Trace each scenario to its root cause\n3. Decide: proceed, modify approach, or reject entirely\n\nThis catches architectural mistakes at the whiteboard, not in production.\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eFirst-Principles Decomposition\u003c/strong\u003e\u003c/summary\u003e\n\nFor Complex tasks, CORTEX prevents agents from pattern-matching to familiar solutions that don't fit. Instead:\n\n1. Strip the problem to its irreducible constraints\n2. Identify the fundamental truths that must hold\n3. Build up from those truths -- not down from existing patterns\n\nThis is the difference between \"how did someone else solve this?\" and \"what does *this problem* actually require?\"\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eIceberg Model\u003c/strong\u003e\u003c/summary\u003e\n\nWhen agents encounter recurring issues, CORTEX forces analysis below the surface:\n\n- **Events:** What happened? *(visible)*\n- **Patterns:** What keeps happening? *(trend)*\n- **Structures:** What system produces this pattern? *(cause)*\n- **Mental models:** What assumptions sustain the structure? *(root)*\n\nFixing events is a band-aid. Fixing mental models prevents entire categories of bugs.\n\n\u003c/details\u003e\n\nThis isn't bolted on as a post-hoc review -- it's wired into the executor's decision loop. Every non-trivial choice goes through CORTEX before code is written.\n\n---\n\n## Commands\n\n### Core Workflow\n\n| Command | What it does |\n|---------|-------------|\n| `/mz:init` | Initialize project: questions, preset selection, planning structure |\n| `/mz:discuss` | Socratic brainstorming to explore approaches before planning |\n| `/mz:plan [N]` | Research + plan + verify for a phase |\n| `/mz:go` | Execute phase plan with Agent Teams coordination |\n| `/mz:verify` | Goal-backward verification against acceptance criteria |\n| `/mz:review` | Two-stage code review (spec compliance + code quality) |\n| `/mz:lifecycle` | Audit milestone, archive, tag release, start next version |\n\n### Navigation\n\n| Command | What it does |\n|---------|-------------|\n| `/mz:status` | Where am I? What's next? |\n| `/mz:help` | Show all commands with usage examples |\n\n### Brownfield\n\n| Command | What it does |\n|---------|-------------|\n| `/mz:map` | Analyze existing codebase (tech, architecture, conventions, concerns) |\n\n### Session Management\n\n| Command | What it does |\n|---------|-------------|\n| `/mz:pause` | Save session context for handoff to future session |\n| `/mz:resume` | Restore context and continue from where you left off |\n\n### Utilities\n\n| Command | What it does |\n|---------|-------------|\n| `/mz:settings` | View and modify project configuration |\n| `/mz:quick` | Execute ad-hoc task with quality guarantees, skip full ceremony |\n| `/mz:debug` | Systematic four-phase debugging (reproduce, isolate, root-cause, fix) |\n\n---\n\n## Why It Works\n\n### Native Agent Teams\n\nMost Claude Code frameworks spawn subagents with the Task tool -- fire and forget. The orchestrator sends work out, collects results, moves on. Agents never talk to each other.\n\nMegazord uses Claude Code's built-in Agent Teams: `TeamCreate`, `SendMessage`, `TaskList`, `TaskUpdate`. Agents coordinate through a shared task list, send messages to each other in real time, and hand off context without losing anything. The orchestrator stays lightweight while agents do the heavy lifting.\n\n| | Agent Teams (Megazord) | Task tool (others) |\n|---|---|---|\n| Communication | Bidirectional via SendMessage | None |\n| Coordination | Shared TaskList/TaskUpdate | Orchestrator manages manually |\n| Agent awareness | Agents see each other's work | Agents work in isolation |\n| Context preservation | Shared state across agents | Each agent starts fresh |\n\n\u003e **Note:** Agent Teams is currently an experimental Claude Code feature. To enable it:\n\u003e ```bash\n\u003e export CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1\n\u003e ```\n\u003e Without this flag, Megazord automatically falls back to subagent mode (still fully functional, just without inter-agent communication).\n\n### CORTEX Adaptive Reasoning\n\nFor how agents reason through complexity, see [CORTEX](#cortex).\n\n### 6 Specialized Agents\n\nEvery agent has one job and does it well:\n\n| Agent | Role |\n|-------|------|\n| **Executor** | Implements tasks, writes code, commits atomically |\n| **Planner** | Creates task breakdown with dependencies and waves |\n| **Researcher** | Investigates technical landscape before planning |\n| **Reviewer** | Two-stage code review (spec compliance + quality) |\n| **Verifier** | Goal-backward verification against acceptance criteria |\n| **Mapper** | Analyzes existing codebases (brownfield support) |\n\n### Context Engineering\n\nQuality degrades as context windows fill up. Megazord keeps things clean:\n\n| File | Purpose |\n|------|---------|\n| `PROJECT.md` | Project vision, always loaded |\n| `ROADMAP.md` | Where you're going, what's done |\n| `STATE.md` | Decisions, blockers, position -- memory across sessions |\n| `PLAN.md` | Task breakdown with dependencies and verification steps |\n| `RESEARCH.md` | Technical landscape for each phase |\n| `CONTEXT.md` | Your implementation decisions from brainstorming |\n| `SUMMARY.md` | What happened, what changed -- committed to history |\n\nEach file is sized to stay in the optimal context range. Fresh agents get exactly the context they need, nothing more.\n\n### Atomic Git Commits\n\nEach task gets its own commit immediately after completion:\n\n```\nabc123f feat(03-01): create user model and migrations\ndef456g feat(03-01): add authentication endpoints\nhij789k feat(03-02): implement password hashing\nlmn012o feat(03-02): create registration flow\n```\n\nGit bisect finds the exact failing task. Each task is independently revertable. Clean history for both humans and AI in future sessions.\n\n### File Ownership Enforcement\n\nWhen multiple agents work in parallel, file conflicts are inevitable -- unless you prevent them. Megazord uses a `PreToolUse` hook that validates file ownership before any `Edit` or `Write` operation. Two agents can never modify the same file simultaneously.\n\n---\n\n## Configuration\n\nMegazord stores project settings in `.planning/megazord.config.json`. Configure during `/mz:init` or update with `/mz:settings`.\n\n### Quality Presets\n\n| Preset | TDD | Review | Brainstorming | Debug | Research | Plan Check | Verifier |\n|--------|-----|--------|---------------|-------|----------|------------|----------|\n| **Strict** | On | Auto | On | Systematic | On | On | On |\n| **Balanced** (default) | Off | Auto | On | Systematic | On | On | On |\n| **Minimal** | Off | Off | Off | Quick | Off | Off | Off |\n\n**Strict** gives you maximum quality gates -- TDD enforced, brainstorming before every plan, systematic debugging. Best for production code where correctness matters.\n\n**Balanced** keeps the important gates (research, verification, review) without the overhead of TDD on every task. Good default for most projects.\n\n**Minimal** strips everything optional. Plan and execute, nothing else. Good for prototyping and throwaway code.\n\n### Model Profiles\n\nControl which Claude model each agent uses. Balance quality against token spend.\n\n| Profile | Planning | Execution | Verification |\n|---------|----------|-----------|--------------|\n| **Quality** | Opus | Opus | Sonnet |\n| **Balanced** (default) | Opus | Sonnet | Sonnet |\n| **Budget** | Sonnet | Sonnet | Haiku |\n\nSwitch profiles:\n\n```\n/mz:settings model_profile quality\n```\n\n### Workflow Toggles\n\nThese control optional agents that improve quality but add tokens and time.\n\n| Setting | Default | What it does |\n|---------|---------|-------------|\n| `workflow.research` | `true` | Research domain before planning each phase |\n| `workflow.plan_check` | `true` | Verify plans achieve phase goals before execution |\n| `workflow.verifier` | `true` | Confirm deliverables after execution |\n\n### Execution Settings\n\n| Setting | Default | What it does |\n|---------|---------|-------------|\n| `mode` | `interactive` | `interactive` confirms at each step, `yolo` auto-approves |\n| `depth` | `standard` | `comprehensive`, `standard`, or `quick` planning thoroughness |\n| `parallelization` | `true` | Run independent tasks simultaneously |\n| `commit_docs` | `true` | Track `.planning/` in git |\n\n---\n\n## Workflow Examples\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eGreenfield: Build a new project from scratch\u003c/strong\u003e\u003c/summary\u003e\n\n\u003e You're starting fresh -- a task management API. Megazord handles the full lifecycle.\n\n```\n/mz:init\n```\n```\nMEGAZORD v1.1.0\nPreset: Balanced\nCreated .planning/megazord.config.json\nCreated .planning/ROADMAP.md\nReady! Run /mz:plan to start planning.\n```\n\n```\n/mz:plan 1\n```\n```\nMEGAZORD \u003e PLANNING\nResearching phase 1...\nCreated PLAN.md (8 tasks, 2 waves)\n```\n\n```\n/mz:go\n```\n```\nMEGAZORD \u003e EXECUTE\nWave 1: Spawning 3 executor agents...\nTask 1.1: Project structure [DONE]\nTask 1.2: Database schema [DONE]\nWave 2: Spawning 2 executor agents...\nAll tasks complete. Run /mz:verify to check deliverables.\n```\n\n```\n/mz:verify\n```\n```\nMEGAZORD \u003e VERIFY\nChecking 4 success criteria...\n[PASS] API endpoints respond correctly\n[PASS] Database migrations run clean\nVerification: PASSED (4/4 criteria met)\n```\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eBrownfield: Add features to an existing codebase\u003c/strong\u003e\u003c/summary\u003e\n\n\u003e You have a SaaS codebase and want to add structured feature work. Start with `/mz:map`.\n\n```\n/mz:map\n```\n```\nMEGAZORD \u003e MAPPING\nScanning project structure...\nFound: Next.js 15, Prisma, 47 routes, 12 models\nPatterns: App Router, server actions, Stripe integration\nMap saved to .planning/codebase/\n```\n\n```\n/mz:plan 1\n```\n```\nMEGAZORD \u003e PLANNING\nUsing codebase map for context...\nCreated PLAN.md (5 tasks, 2 waves)\nPlans account for existing patterns and conventions.\n```\n\n```\n/mz:go\n```\n```\nMEGAZORD \u003e EXECUTE\nWave 1: Spawning 2 executor agents...\nTask 1.1: Add user dashboard route [DONE]\nTask 1.2: Extend Prisma schema [DONE]\nAll tasks complete.\n```\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eQuick task: Skip the ceremony\u003c/strong\u003e\u003c/summary\u003e\n\n\u003e Small change, full planning would be overkill.\n\n```\n/mz:quick \"Add dark mode toggle to settings page\"\n```\n```\nMEGAZORD \u003e QUICK\nAnalyzing task scope...\nCreating settings/theme-toggle.tsx\nUpdating settings/page.tsx\nAdding dark mode CSS variables to globals.css\nDone. 3 files modified, 1 created.\n```\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eDebug: Track down a tricky bug\u003c/strong\u003e\u003c/summary\u003e\n\n\u003e Stale cart totals after adding items. Let Megazord work through it systematically.\n\n```\n/mz:debug \"Users see stale cart totals after adding items\"\n```\n```\nMEGAZORD \u003e DEBUG\nPhase 1: Reproducing...\n  Found: cart total updates on refresh but not on add\nPhase 2: Isolating...\n  Narrowed to: useCart hook not invalidating query cache\nPhase 3: Root cause...\n  Cache key mismatch between addItem mutation and cart query\nPhase 4: Fixing...\n  Updated mutation to invalidate ['cart', userId] query key\nFix applied and verified. 1 file modified.\n```\n\n\u003c/details\u003e\n\n\u003cdetails\u003e\n\u003csummary\u003e\u003cstrong\u003eSession continuity: Pick up where you left off\u003c/strong\u003e\u003c/summary\u003e\n\n\u003e End of day. You're mid-phase and don't want to lose context.\n\n```\n/mz:pause\n```\n```\nMEGAZORD \u003e PAUSE\nCaptured: Phase 3, 4/7 tasks complete\nNext: Task 3.5 (Add email notification service)\nStashed 2 modified files\nContext saved to STATE.md\n```\n\nNext session:\n\n```\n/mz:resume\n```\n```\nMEGAZORD \u003e RESUME\nRestored: Phase 3, task 3.5 pending\nUnstashed 2 files\nContext loaded. Ready to continue.\n```\n\n\u003c/details\u003e\n\n---\n\n## What's Inside\n\n### Project Structure\n\n```\n~/.claude/megazord/                  # Plugin installation\n├── .claude-plugin/plugin.json       # Plugin manifest\n├── skills/                          # 15 slash commands\n│   ├── init/SKILL.md\n│   ├── plan/SKILL.md\n│   ├── go/SKILL.md\n│   ├── verify/SKILL.md\n│   └── ...\n├── agents/                          # 6 specialized agents\n│   ├── mz-executor.md\n│   ├── mz-planner.md\n│   ├── mz-researcher.md\n│   ├── mz-reviewer.md\n│   ├── mz-verifier.md\n│   └── mz-mapper.md\n├── hooks/hooks.json                 # File ownership enforcement\n└── scripts/enforce-ownership.sh     # PreToolUse safety hook\n```\n\n### Your Project (after `/mz:init`)\n\n```\nyour-project/\n├── .planning/\n│   ├── megazord.config.json         # Project configuration\n│   ├── PROJECT.md                   # Vision, goals, requirements\n│   ├── STATE.md                     # Current position, decisions\n│   ├── ROADMAP.md                   # Phase breakdown, success criteria\n│   ├── phases/                      # Per-phase planning artifacts\n│   │   └── 01-setup/\n│   │       ├── CONTEXT.md           # Brainstorm decisions\n│   │       ├── RESEARCH.md          # Technical research\n│   │       ├── PLAN.md              # Task breakdown + waves\n│   │       ├── SUMMARY.md           # Execution results\n│   │       └── VERIFICATION.md      # Verification report\n│   └── codebase/                    # Brownfield analysis (from /mz:map)\n│       ├── STACK.md\n│       ├── ARCHITECTURE.md\n│       ├── CONVENTIONS.md\n│       └── SUMMARY.md\n└── your code...\n```\n\n---\n\n## Troubleshooting\n\n**Commands not found after install?**\n- Restart Claude Code to reload slash commands\n- Run `bunx megazord-cli` again to reinstall\n\n**Commands not working as expected?**\n- Run `/mz:help` to verify installation\n- Check `/mz:status` for project state\n\n**Updating to the latest version?**\n\n```bash\nbunx megazord-cli update\n```\n\n**Uninstalling?**\n\n```bash\nbunx megazord-cli uninstall\n```\n\nThis removes the plugin while preserving your project's `.planning/` directory.\n\n---\n\n## License\n\n[MIT](LICENSE) -- use it, fork it, build on it.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsh3rd3n%2Fmegazord","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsh3rd3n%2Fmegazord","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsh3rd3n%2Fmegazord/lists"}