nexgsd
v1.4.3
Published
From one prompt to a finished product — fully autonomous. NexGSD turns any AI coding tool into a complete project engine: auto-plans, remembers across sessions, builds end-to-end, audits security & performance, deploys to production. 18 agents + 39 workfl
Downloads
557
Readme
NexGSD
From one prompt to a finished product — fully autonomous.
NexGSD turns any AI coding tool into a complete project engine. It plans your project, remembers everything between sessions, builds phase by phase, audits security and performance, and ships to production — while you watch (or walk away and get notified on your phone).
npm i -g nexgsd && nexgsd installWhat NexGSD does for you
- Auto-planning — give it a prompt like "build a SaaS dashboard with auth and billing", it researches the domain, writes requirements, creates a phased roadmap. No spec needed.
- Persistent memory — a
.planning/directory remembers every decision, every phase, every context across sessions. Your AI never forgets where it left off. - End-to-end builds from scratch — plan → research → build with atomic commits → verify → audit → deploy, zero babysitting. Walk away, come back to a shipped product.
- Anti-hallucination guardrails — 8 structural safeguards: file-first reads, source verification, confidence tagging, no auto-pass on tests.
- Quality audits built-in — security (OWASP), performance (Lighthouse), accessibility (WCAG 2.2), mobile responsiveness, SEO, brand consistency — all run in parallel.
- Push notifications — ntfy.sh integration pings your phone when phases complete, deploys succeed, or something needs your attention.
- 18 specialized agents + 39 workflows — from
/nexgsd-new-project(planning) to/nexgsd-super(full autonomy) to/nexgsd-deploy(production).
Works with Claude Code, GitHub Copilot CLI, OpenAI Codex CLI, Cursor, Windsurf, Cline, Antigravity, and any AI coding tool that supports custom agents or slash commands.
"I built this because every AI coding tool is great at writing code but terrible at managing projects. NexGSD bridges that gap." — Mert Ali Dalkır, maintainer
Built at NexVar — the execution framework behind everything we ship.
Why NexGSD?
The problem: AI coding assistants write code fast but lose context between sessions, hallucinate project state, skip testing, forget to audit security, and can't manage multi-phase projects.
The solution: NexGSD adds a structured execution layer on top of any AI coding tool:
- Context management —
.planning/directory persists project state across sessions - Anti-hallucination — every claim verified against real files and real command output
- Quality gates — security, performance, mobile, SEO, accessibility, brand audits built-in
- Autonomous execution — give a prompt, walk away, get notified when done
- Phased delivery — complex projects broken into manageable phases with atomic commits
Quick Start
Option A: npm (recommended)
npm i -g nexgsd
cd your-project
nexgsd installThat's it. Works on macOS, Linux, and Windows.
One-time use without global install:
cd your-project
npx nexgsd installUpdate later:
npm update -g nexgsd
nexgsd updateOption B: Git clone (manual)
git clone https://github.com/NexVar/NexGSD.gitmacOS / Linux / WSL:
bash NexGSD/nexgsd-install.sh /path/to/your-projectWindows (PowerShell):
# Use Git Bash
bash NexGSD/nexgsd-install.sh C:/Users/you/your-project
# Or manual copy
Copy-Item -Recurse NexGSD/.agent your-project/.agentStart building
Open any AI coding tool (Claude Code, Copilot, Codex, Cursor, Antigravity, etc.) and follow this flow:
Step 1 — Plan your project:
/nexgsd-new-projectIt will ask you questions about what you want to build, research the domain, and create a structured project plan. No code is written at this stage.
Step 2 — Build it autonomously:
/nexgsd-superAfter planning, run this to execute the entire project autonomously — plan, build, test, audit, and deploy. Walk away and get notified when it's done.
Or skip planning and go full autonomous with a prompt:
/nexgsd-super "Build a SaaS dashboard with auth, billing, and analytics"See all 39 commands:
/nexgsd-helpWorks with any AI coding tool. For non-slash-command tools, just tell them: "Read
.agent/workflows/nexgsd-new-project.mdand follow the workflow."
CLI Commands
nexgsd install [path] # Install NexGSD to a project
nexgsd update [path] # Update to latest version
nexgsd info # Show agent/workflow counts
nexgsd --version # Show version
nexgsd --help # Show helpUsage by Platform
Claude Code (CLI)
NexGSD is natively designed for Claude Code. All workflows are slash commands:
cd your-project
claude
/nexgsd-new-project # Initialize project
/nexgsd-plan 1 # Plan phase 1
/nexgsd-execute 1 # Execute with atomic commits
/nexgsd-audit # Full quality audit
/nexgsd-deploy # Deploy to productionGitHub Copilot CLI / OpenAI Codex CLI
The .agent/ directory works as context files. Point the CLI to read them:
# Copilot CLI
copilot "Read .agent/workflows/nexgsd-new-project.md and follow the workflow to initialize this project"
# Codex CLI
codex "Follow the workflow in .agent/workflows/nexgsd-super.md to build: [your prompt]"Add a .github/copilot-instructions.md referencing NexGSD:
When I say /nexgsd-[command], read and follow .agent/workflows/nexgsd-[command].md
Available agents are in .agent/agents/Cursor / Windsurf / Cline / VS Code (IDE Agents)
- Install NexGSD to your project:
./nexgsd-install.sh . - Open your project in the IDE
- Reference workflows in chat:
- "Follow .agent/workflows/nexgsd-new-project.md to set up this project"
- "Use the nexgsd-security-auditor agent to scan for vulnerabilities"
- The IDE agent reads the
.agent/files as context and follows the structured workflows
Any LLM / Custom Setup
NexGSD agents and workflows are plain Markdown files. Any LLM that can read files and execute commands can use them:
System prompt: "You have access to a project management framework in .agent/.
Read .agent/workflows/nexgsd-help.md for available commands. When the user requests
a workflow, read and follow the corresponding .md file step by step."Default Workflow (The NexGSD Loop)
This is how most projects flow from idea to production:
┌─────────────────────────────────────────────────────────────┐
│ THE NEXGSD LOOP │
│ │
│ ┌──────────────┐ │
│ │ /nexgsd-new-project │ ← You describe what to build │
│ └──────┬───────┘ │
│ │ Creates: PROJECT.md, REQUIREMENTS.md, │
│ │ ROADMAP.md, STATE.md, nexgsd-config.json │
│ ▼ │
│ ┌──────────────┐ │
│ │ /nexgsd-plan N │ ← Research + create task plans │
│ └──────┬───────┘ │
│ │ Creates: RESEARCH.md, PLAN.md files │
│ ▼ │
│ ┌──────────────┐ │
│ │ /nexgsd-execute N│ ← Build with atomic commits │
│ └──────┬───────┘ │
│ │ Creates: SUMMARY.md, VERIFICATION.md │
│ │ Sends: push notification ✓ │
│ ▼ │
│ ┌──────────────┐ │
│ │ /nexgsd-verify N│ ← User acceptance testing │
│ └──────┬───────┘ │
│ │ │
│ ├── Pass? → Next phase (repeat from /nexgsd-plan N+1) │
│ └── Gaps? → /nexgsd-plan N --gaps (fix and re-execute) │
│ │
│ After all phases: │
│ ┌──────────────┐ │
│ │ /nexgsd-audit │ ← Security, performance, mobile, SEO, │
│ └──────┬───────┘ accessibility, brand review │
│ ▼ │
│ ┌──────────────┐ │
│ │ /nexgsd-deploy │ ← Build, deploy, verify live URL │
│ └──────────────┘ │
│ │
│ SHORTCUT: /nexgsd-super does ALL of this autonomously │
└─────────────────────────────────────────────────────────────┘For quick tasks that don't need full project setup:
/nexgsd-quick "Add dark mode toggle to the navbar"Command Surface
Core
| Command | What it does |
|---------|-------------|
| /nexgsd-super [prompt] | Full autonomy — prompt to production, zero human input |
| /nexgsd-new-project | Interactive setup: questions → research → requirements → roadmap → config |
| /nexgsd-no-halluc [question] | Verified Q&A with mandatory external research and citations |
Build Cycle
| Command | What it does |
|---------|-------------|
| /nexgsd-discuss [phase] | Capture implementation decisions before planning |
| /nexgsd-plan [phase] | Research and create executable task plans |
| /nexgsd-execute [phase] | Execute plans with atomic git commits |
| /nexgsd-verify [phase] | User acceptance testing against phase goals |
Quality & Review
| Command | What it does |
|---------|-------------|
| /nexgsd-audit | Full project audit — 6 agents in parallel |
| /nexgsd-review [phase] | PR-style code review with multi-agent analysis |
| /nexgsd-refactor [desc] | Safe refactoring with test snapshots + auto-rollback |
Operations
| Command | What it does |
|---------|-------------|
| /nexgsd-deploy | Build → deploy → verify (CF Pages / Vercel / Netlify) |
| /nexgsd-migrate [desc] | DB migration with safety gates + rollback SQL |
| /nexgsd-setup-config | Configure ntfy, deploy platform, Supabase, tests |
Utilities
| Command | What it does |
|---------|-------------|
| /nexgsd-quick [desc] | Small ad-hoc task with NexGSD guarantees |
| /nexgsd-progress | Current state, blockers, next steps |
| /nexgsd-commit-memory | Distill context into long-term memory |
| /nexgsd-help | Show all commands |
Brownfield & Milestones
| Command | What it does |
|---------|-------------|
| /nexgsd-map-codebase | Build a real codebase map before planning existing projects |
| /nexgsd-new-milestone | Start the next milestone on an existing project |
| /nexgsd-audit-milestone | Verify milestone-wide requirement coverage and integration |
| /nexgsd-complete-milestone | Archive a shipped milestone and prepare the next cycle |
| /nexgsd-plan-milestone-gaps | Turn milestone audit gaps into roadmap work |
Continuity & Maintenance
| Command | What it does |
|---------|-------------|
| /nexgsd-debug [issue] | Start or resume a structured debug session |
| /nexgsd-pause-work | Save a precise handoff for the next session |
| /nexgsd-resume-work | Restore project context and route to the next step |
| /nexgsd-add-todo | Capture follow-up work without losing context |
| /nexgsd-check-todos | Review pending todos and route them into action |
| /nexgsd-health | Validate .planning/ integrity and repair safe issues |
| /nexgsd-update | Update a NexGSD clone or explain reinstall path |
| /nexgsd-settings | Inspect or update NexGSD project settings |
| /nexgsd-set-profile | Set the preferred model profile |
| /nexgsd-research-phase | Run standalone phase research before planning |
| /nexgsd-add-phase | Append a new roadmap phase |
| /nexgsd-insert-phase | Insert an urgent decimal phase |
| /nexgsd-remove-phase | Remove an unstarted future phase |
| /nexgsd-add-tests | Add or strengthen tests for an area or phase |
| /nexgsd-reapply-patches | Reapply locally saved patch overlays after updates |
All Agents (18)
Planning & Execution
| Agent | What it does |
|-------|-------------|
| nexgsd-executor | Executes plans with atomic commits, deviation handling, checkpoints |
| nexgsd-planner | Creates task plans with dependency analysis and wave ordering |
| nexgsd-roadmapper | Creates phased project roadmaps from requirements |
| nexgsd-phase-researcher | Researches implementation approach before planning |
| nexgsd-project-researcher | Researches domain, stack, ecosystem before roadmap |
| nexgsd-research-synthesizer | Merges parallel research outputs into actionable summaries |
Verification
| Agent | What it does |
|-------|-------------|
| nexgsd-verifier | Goal-backward verification — did the code deliver what was promised? |
| nexgsd-plan-checker | Pre-execution plan quality check |
| nexgsd-integration-checker | Cross-phase integration and E2E flow verification |
| nexgsd-debugger | Scientific method bug investigation with hypothesis testing |
| nexgsd-codebase-mapper | Explores and documents codebase structure |
Quality Auditors
| Agent | What it does |
|-------|-------------|
| nexgsd-security-auditor | OWASP top 10, dependency audit, secrets detection, CSP |
| nexgsd-performance-tester | Lighthouse, bundle size, lazy loading, render performance |
| nexgsd-mobile-auditor | Responsive design, touch targets, viewport, overflow |
| nexgsd-seo-checker | Metadata, sitemap, structured data, hreflang, headings |
| nexgsd-accessibility-tester | WCAG 2.2 AA, ARIA, keyboard nav, contrast, focus |
| nexgsd-brand-reviewer | Brand consistency, copy quality, design system, typography |
Infrastructure
| Agent | What it does |
|-------|-------------|
| nexgsd-notifier | Push notifications via ntfy.sh at milestones |
Anti-Hallucination System
The #1 problem with AI coding: it says it did something but didn't actually do it. NexGSD has 8 structural safeguards:
| Protection | How |
|------------|-----|
| File-First Context | Always re-reads files before acting — never trusts memory |
| Source Verification | Technical claims verified against docs, not training data |
| Confidence Levels | Research tagged HIGH (verified) / MEDIUM (searched) / LOW (memory) |
| Verification Gates | Every task output is READ — not assumed to pass |
| No Auto-Pass | Build/test results never marked passed without actual output |
| Decision Attribution | Tracks USER-decided vs AI-suggested for audit trail |
| Context Freshness | New conversation recommended between workflow steps |
| Checkpoint Integrity | Waits for human approval — never skips confirmation |
Push Notifications
Configure during /nexgsd-new-project or /nexgsd-setup-config:
"Want push notifications?" → yes
"ntfy channel name?" → my-project-alertsYou'll get notified on your phone when:
- A phase completes
- Deploy succeeds or fails
- Audit finds critical issues
- Errors block execution
Uses ntfy.sh — free, open-source, no account needed.
File Structure
your-project/
├── .agent/ ← NexGSD system (copied by install)
│ ├── agents/ ← 18 specialized agents
│ │ ├── nexgsd-executor.md
│ │ ├── nexgsd-planner.md
│ │ ├── nexgsd-security-auditor.md
│ │ └── ... (18 total)
│ └── workflows/ ← 39 workflows
│ ├── nexgsd-new-project.md
│ ├── nexgsd-super.md
│ ├── nexgsd-audit.md
│ └── ... (39 total)
│
└── .planning/ ← Project state (created by nexgsd-new-project)
├── PROJECT.md ← Vision and context
├── REQUIREMENTS.md ← v1/v2 requirements
├── ROADMAP.md ← Phases and progress
├── STATE.md ← Current position (living memory)
├── nexgsd-config.json ← Settings (ntfy, deploy, tests)
├── research/ ← Domain research
└── phases/
└── 01-phase-name/
├── 01-01-PLAN.md ← Task plan
├── 01-01-SUMMARY.md ← Execution summary
├── 01-VERIFICATION.md ← Goal verification
├── SECURITY-AUDIT.md ← Security report
├── PERFORMANCE-REPORT.md
├── MOBILE-AUDIT.md
├── SEO-REPORT.md
├── ACCESSIBILITY-REPORT.md
├── BRAND-REVIEW.md
└── AUDIT.md ← Unified auditReal-World Example
This system was built while shipping real production projects:
- NexVar.io — AI-first software studio site (Next.js 16, i18n, AI chatbot, newsletter, blog CMS)
- Vault — Personal finance tracker (Supabase, multi-currency, AI assistant, real-time rates)
Every agent and workflow exists because it was needed during real development. Read more at mertdlkr.com.
Requirements
Required:
- Git
- Any AI coding CLI or IDE (Claude Code, Copilot, Codex, Cursor, Windsurf, Cline)
Optional:
- Node.js 18+ (for web projects)
- ntfy app (push notifications)
- Supabase / Prisma (database features)
- Cloudflare / Vercel / Netlify (deployment)
Model Compatibility
NexGSD works with any LLM:
| Model | Status | |-------|--------| | Claude (Anthropic) | Fully tested | | GPT / Codex (OpenAI) | Compatible | | Gemini (Google) | Compatible | | Llama / Mistral / DeepSeek | Compatible |
Model-agnostic by design — uses structural safeguards (file reads, command verification, user gates) instead of model-specific prompting.
Contributing
PRs welcome. If you add a new agent or workflow, follow the existing format in .agent/agents/ and .agent/workflows/.
Built By
NexVar
AI-first software studio shipping production tooling, AI platforms, and developer infrastructure. NexGSD is the execution framework we use internally on every project we ship.
Mert Ali Dalkır
Creator and maintainer. Co-founder of NexVar. AI-first builder making AI-powered development faster, more reliable, and more structured.
License
MIT © NexVar and Mert Ali Dalkır
If NexGSD saves you time, give it a star. It helps others find it.
Built with conviction at nexvar.io and mertdlkr.com.
