By potenlab
Describe an app in plain language, get a running app + PR. A designer or PM speaks product; Claude does the full-stack engineering underneath (data model -> backend -> UI -> running app -> PR) and only ever shows a working app, never code. Auto-triggers on app-creation intent; builds inside an isolated sandbox that never touches production data, secrets, or auth/billing.
Initialize the Designer Workflow in the current project — set where sandbox apps live, draw the production isolation boundary, check required tools/MCPs, and confirm the design + supabase-integration skills are ready. Run this once per project after installing the plugin.
Plan an app or feature before building it — clarify the idea with questions, write a technical spec to /docs/plan, and open a designer-language GitHub epic + child-story issues, then (on your okay) build it autonomously story by story. The explicit way to kick off the planning pipeline when auto-trigger is borderline.
Build ONE app or story full-stack from an already-agreed plan — backend, UI, assets, and wiring inside an isolated sandbox branch, then run it and open a PR. Normally called by goal-loop once per child story; use directly only for a single explicit one-off build. For any fresh change request — building OR editing/restyling/fixing — use plan-and-spec FIRST, not this. Responds only in plain product language and a working app, never raw code.
Use after plan-and-spec to build a planned change to completion. Scales to size: a small single-issue change is built directly (one story, verify, one PR); a new/large epic is walked one child story at a time (build via design, verify in a browser, open one PR per story), then reports. Triggered by plan-and-spec ONLY after the user has confirmed the FILED GitHub issue is correct; or use directly when a spec (/docs/plan) and a filed, user-approved issue already exist. Execution only — it does not plan or file issues.
Use whenever a visual or media ASSET is needed for an app or page — an image, icon, logo, illustration, hero/background, avatar, texture, video clip, sound effect, or voiceover. Higgsfield is the ONE path for generating assets: never use stock photos, emoji, placeholder boxes, hand-drawn SVGs, or another generator, and never ask the user to go find assets. AUTO-TRIGGER on asset intent: 'add an image/icon/logo/illustration', 'needs a hero', 'a background for…', 'generate a picture/video/sound', or any build step that requires visual/media content. Scope is assets ONLY — do NOT use this for code, data, schema, copywriting, or UI layout.
You MUST use this before ANY change to the project's code or UI — building something new, OR editing, restyling, tweaking, fixing, or refactoring anything that already exists — before you touch any file. Clarifies the change through questions, writes a technical spec to /docs/plan, and ALWAYS files a GitHub issue with gh, then hands off to goal-loop to build. Skip ONLY for pure questions, explanations, or read-only code review with no change.
Wire a Supabase backend into a freshly created app the right way the first time — schema, RLS, auth, migrations, and type-safe client — driven by the Supabase MCP. Use when the /design app-creation flow (or any task) needs to give a new app a working, secure Postgres backend. Triggers: 'add a backend', 'store data', 'save submissions', 'log in', 'user accounts', 'database', Supabase, RLS, migrations, auth, edge functions. Respond to designers/PMs in plain language; do the engineering underneath.
External network access
Connects to servers outside your machine
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. A designer or PM describes an application in plain language and gets a running app + a PR — Claude does the full-stack engineer's thinking underneath (idea → data model → backend → UI → running app → PR) but only ever responds in design/product language and a working app, never code. It auto-triggers on intent: no command to memorize, no diffs to read.
See docs/TIMELINE.md for the full spec, locked decisions, and acceptance
criteria. (docs/proposal.html / docs/proposal.pdf are the shareable versions.)
1 — Install the plugin in Claude Code:
/plugin marketplace add potenlab/designer-workflow
/plugin install designer-workflow@designer-workflow
(or non-interactively: claude plugin marketplace add potenlab/designer-workflow then
claude plugin install designer-workflow@designer-workflow.)
2 — Initialize it in your project — run once per project:
/designer-workflow:dw-init
dw-init asks, in plain language, where new apps should live and what's off-limits (your real
data, secrets, auth/billing), checks the tools the workflow needs (Supabase MCP, a browser tool,
optionally higgsfield), and writes .designer-workflow/config.md. After that there is no command —
just describe an app and the workflow takes over.
Just say what you want:
"a clean little gallery app where I drop images and tag them" "an intake tool: clients submit a brief, and we track it New → Doing → Done"
It runs on any change — building something new, or editing/restyling/fixing something that exists — and runs the pipeline plan → file issue → confirm → build → ship:
plan-and-spec) — asks a few plain-language questions (or none for a small
change — it scales to size) and writes a technical spec to docs/plan/ (for the developer, never
shown in chat).gh: a single tracking
issue for a small change, or an epic + one child issue per piece for new/large work, in plain
product language.goal-loop) — on your yes: a small change is built directly (verify → one PR); new/
large work is built one piece at a time, each on an isolated sandbox branch, with a mobile +
desktop walkthrough and one PR per piece that closes its issue. Never touches other apps,
production data, secrets, or auth/billing.You make one decision — confirming the filed issue; after that it runs to done. Say "just file it,
don't build yet" to stop at the issue. There's no command to remember, but /designer-workflow:dw-plan
kicks off planning explicitly if you want to be certain.
The split of audiences is deliberate: chat stays plain language, the technical spec lives in
docs/plan/, and the GitHub issues are designer/product language for the team to track.
Full-stack project planning and execution plugin for Claude Code. Orchestrates specialist agents to plan, build, and iterate on projects.
Dual knowledge-graph workflow built on understand-anything: Impl KG (autoUpdate, derived from source) + SSOT KG (intent/spec, projected from a Decision Log). Bootstraps via /kg-init.
Agentic SaaS validation flow: brand intake → Higgsfield logo + moodboard → approval gate → animated landing page with waitlist. Powered by the Higgsfield MCP server.
npx claudepluginhub potenlab/designer-workflow --plugin designer-workflowHarness-native ECC plugin for engineering teams - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Persistent file-based planning for AI coding agents. Crash-proof markdown plans (task_plan.md, findings.md, progress.md) that survive context loss and /clear, with an opt-in completion gate and multi-agent shared state. Manus-style. Works with Claude Code, Codex CLI, Cursor, Kiro, OpenCode and 60+ agents via the SKILL.md standard. Includes Arabic, German, Spanish, and Chinese (Simplified and Traditional).
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
v9.44.1 — Patch release for Gemini environment/version detection and qwen auth gating. Run /octo:setup.
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.