Help us improve
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
Share bugs, ideas, or general feedback.
STUDIO - Autonomous software development workflow: Plan then Build with embedded context and quality gates
npx claudepluginhub twofoldtech-dakota/studio --plugin studioCreate strategic, brand-aligned blog content through guided workflow
Establish and manage brand identity, voice, and messaging through guided discovery
Execute approved plans through iterative build loop with quality gates and learning capture
Explicitly coordinate multi-agent workflows with visibility and control
Show project dashboard with progress tracking
Share bugs, ideas, or general feedback.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge.
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
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.
An agent-routed harness for end-to-end software product development
Persona-driven AI development team: orchestrator, team agents, review agents, skills, slash commands, and advisory hooks for Claude Code
AI-powered development workflow automation - Phase-based planning, implementation orchestration, preflight code quality checks with security scanning, ship-it workflow, and development principles generator for CLAUDE.md
Describe your goal, approve the spec, then step away — Claude and Codex loop together until it's right.
End-to-end development workflow: design → draft-plan → orchestrate → review → pr-create → pr-review → pr-merge
Analyze Optimizely CMS/Content Cloud projects for architecture, security, content modeling, and performance patterns
Claude Code plugin for Sitecore, Umbraco, and Optimizely CMS development — scaffolding, content modeling, accessibility scanning, security audits, best practices, and code generation
Analyze Umbraco 14-16 projects for architecture, security, and backoffice patterns
Analyze Optimizely Experimentation and Web implementations for A/B testing, feature flags, and JavaScript SDK patterns
Sitecore XP AWS-to-Azure migration planning assistant. Guides architects through structured discovery, cross-reference analysis, phased estimation, and client-ready deliverable generation.
AI that plans before it builds, learns from mistakes, and never forgets.
Quick Start • How It Works • Commands • Scripts • Full Guide
Most AI coding assistants:
STUDIO fixes this.
# Install
/plugin marketplace add https://github.com/twofoldtech-dakota/studio.git
/plugin install studio@twofoldtech-dakota
# Plan something
/studio "Add user authentication with email verification"
# Build the plan
/build task_xxx
That's it. STUDIO asks questions, creates a plan, and executes with validation.
YOU STUDIO OUTPUT
│ │ │
│ "Add auth" │ │
├───────────────────────►│ │
│ │ │
│ ◄── Questions ───────┤ (3 rounds) │
│ ─── Answers ────────►│ │
│ │ │
│ ◄── Plan ────────────┤ (review & approve) │
│ ─── "looks good" ───►│ │
│ │ │
│ ├───────────────────────►│ Code
│ │ Build + Validate │ Tests
│ │ Quality Gates │ Docs
│ │ Learn & Remember │
│ │ │
| Phase | What Happens | Script |
|---|---|---|
| Plan | Questions → Requirements → Steps | confidence-score.sh |
| Validate | Structure check, confidence ≥70 | validate-plan.sh |
| Pre-check | Lint, types, existing issues | quality-precheck.sh |
| Build | Execute steps, track progress | step-progress.sh |
| Verify | Acceptance criteria, DoD | verify-ac.sh, dod-check.sh |
| Learn | Capture patterns, evolve knowledge | sprint-evolution.sh |
| Command | Alias | What it does |
|---|---|---|
/studio "goal" | /s | Start planning with questions |
/build task_xxx | /b | Execute an approved plan |
/build --resume | Continue from last step | |
/status | Check current task state |
See docs/QUICK-REFERENCE.md for all commands.
All scripts output JSON, have --help, and live in scripts/.
# Before build
./scripts/validate-plan.sh --task-id task_xxx # Structure check
./scripts/confidence-score.sh --task-id task_xxx # Quality score (0-100)
./scripts/quality-precheck.sh # Lint + typecheck
# During build
./scripts/step-progress.sh status task_xxx # Track progress
./scripts/error-matcher.sh --input "error text" # Fix suggestions
# After build
./scripts/verify-ac.sh --task-id task_xxx # Acceptance criteria
./scripts/dod-check.sh --auto-detect # Definition of Done
Plans are scored 0-100:
| Category | Points | Checks |
|---|---|---|
| Requirements | 25 | User confirmations, edge cases, scope |
| Step Quality | 25 | Success criteria, atomic actions |
| Context | 25 | Constraints, quality requirements |
| Risk | 25 | Retry behavior, failure handling |
≥85 = PROCEED • 70-84 = CAUTION • <70 = BLOCKED
STUDIO learns from every build:
| Section | Purpose |
|---|---|
| Strict Constraints | Rules that must never be violated |
| Slop Ledger | Naming/structural mistakes to avoid |
| Performance Delta | Measured improvements with metrics |
| Pending Queue | Signals awaiting promotion |
Every 5 tasks → automatic evolution proposals.