Idea Superpowers Suite logo

Idea Superpowers Suite

Community
AkoliteZA
idea-superpowers-suite

Use when running the full idea workflow: capture a rough idea, expand it into design/UI/implementation docs, research similar products, and generate build-ready Markdown artifacts.

Overview

PublisherAkoliteZA
Repositoryhermes-agent-idea-workflow
Skill nameidea-superpowers-suite
Stars
275
Forks
24
Bundled files
5
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 5 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by AkoliteZA on GitHub. Read the source before you install it.

Installation

Install the Idea Superpowers Suite AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/AkoliteZA/hermes-agent-idea-workflow/tree/main/idea-superpowers-suite
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/AkoliteZA/hermes-agent-idea-workflow.git /tmp/hermes-agent-idea-workflow
mkdir -p .claude/skills
cp -r /tmp/hermes-agent-idea-workflow/idea-superpowers-suite .claude/skills/idea-superpowers-suite
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Idea Superpowers Suite in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Idea Superpowers Suite on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Idea Superpowers Suite is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Idea Superpowers Suite

This skill is a superpowers-style umbrella workflow for idea development.

It coordinates the focused stages:

  • idea-to-design-doc for product/design thinking
  • idea-to-ui-design-brief for optional Full-mode UI direction and image-generation concepts
  • idea-to-implementation-doc for build-ready implementation planning

Use this skill when the user wants a bigger, multi-step workflow that feels like a reusable system rather than a one-off prompt.

Purpose

Turn a rough thought into a structured chain of artifacts:

  1. Idea capture — create or update a local idea note from the user's high-level concept.
  2. Clarifying interview — capture the rough plan, philosophy, target experience, and constraints.
  3. Design doc — flesh the idea into a full product/UX/technical design document.
  4. Optional UI design brief — in Full mode only, define screens, layout, visual direction, states, and optional image-generation concept prompts.
  5. Research pass — compare similar products and identify what to emulate or avoid.
  6. Implementation spec — translate the design into practical build strategy.
  7. Build handoff — produce one Markdown file with tasks, prompts, tests, acceptance criteria, and Superpowers handoff guidance.
  8. Spec review — judge readiness before execution.

Design Philosophy

This workflow should:

  • capture fast;
  • clarify deeply;
  • research before building;
  • separate product thinking from engineering thinking;
  • separate UI direction from implementation details when interface quality matters;
  • save every stage as its own Markdown artifact;
  • remain skippable and resilient when the user wants speed.

The UI brief is optional. It improves interface-heavy projects, but it must not make Lite mode heavier or break existing design-doc → implementation-doc flow.

Mode Selection

At the start, choose one of two modes unless the user explicitly picks one:

  • Lite mode — for sketches, small utilities, early thoughts, and ideas the user wants captured quickly.
  • Full mode — for serious app/product ideas that may later be built by an AI coding agent or Superpowers.

Default to Lite mode when the idea is vague or exploratory. Default to Full mode when the user asks for a product spec, implementation plan, build handoff, research pass, UI/mockup direction, or agent-ready artifact.

Default File Layout

Use local Markdown files first. If the user later wants Obsidian export, treat that as a separate step.

Lite mode layout

text
ideas/<idea-slug>.md
ideas/index.md

Optional only if requested:

text
ideas/<idea-slug>.implementation.md
ideas/<idea-slug>.build-prompt.md
ideas/<idea-slug>.ui-brief.md

Full mode layout with UI stage

text
ideas/<idea-slug>/
  README.md
  00-idea-capture.md
  01-design-doc.md
  02-ui-design-brief.md
  03-implementation-spec.md
  04-agent-build-handoff.md
  05-spec-review.md

Full mode fallback layout when UI stage is skipped

If the UI stage is skipped, preserve the previous simpler numbering:

text
ideas/<idea-slug>/
  README.md
  00-idea-capture.md
  01-design-doc.md
  02-implementation-spec.md
  03-agent-build-handoff.md
  04-spec-review.md

Use whichever numbering is already established in a package. Keep the README artifact map consistent.

Workflow Stages

Stage 1: Capture

When the user says they have an idea, create a working note with:

  • title;
  • short summary;
  • rough problem statement;
  • initial bullet points;
  • mode: Lite or Full.

If needed, ask for a name, but prefer to proceed with a temporary title and refine later.

Stage 2: Interview

Ask one question at a time. Use references/interview-question-bank.md for stronger prompts.

In Lite mode, ask only enough to capture the idea clearly, then draft.

In Full mode, cover:

  • target user and problem;
  • what the app should do;
  • desired behavior and product experience;
  • scope and non-goals;
  • main screens or sections;
  • what should be visible at a glance;
  • where users take action;
  • details that can be hidden until needed;
  • desired visual taste: minimal, playful, professional, dense, dashboard-like, calm, command-center, etc.;
  • empty, loading, error, success, first-run, and permission states where relevant;
  • data, integrations, and platform constraints;
  • data location and hosting preference;
  • platform targets: web, desktop, mobile web, iOS, Android, etc.;
  • topology: one app or multiple pieces such as desktop app + hosted web app + API + worker;
  • authentication, secrets, API keys, public/private sharing, and client/server boundaries;
  • what would make the result feel excellent, not merely functional.

Do not push into stack decisions too early unless the user has made a technical constraint explicit. When technical planning starts, recommend practical defaults first, then let the user accept or change them.

Stage 3: Design Doc

Write 01-design-doc.md to capture product direction and enough technical shape to support planning.

Suggested sections:

  • one-line summary;
  • problem / purpose;
  • product philosophy;
  • target user;
  • core concept;
  • desired behavior;
  • key features;
  • layout / information architecture;
  • UX notes;
  • technical shape;
  • data / integrations / platform needs;
  • hosting / data location / deployment preference;
  • platform targets;
  • recommended technical defaults and accepted/changed decisions;
  • non-goals;
  • open questions;
  • next steps.

Stage 4: Optional UI Design Brief

This stage is Full-mode only by default. Do not add it to Lite mode unless the user explicitly asks for UI, mockups, generated images, or visual design direction.

Load idea-to-ui-design-brief when this stage runs.

Create 02-ui-design-brief.md when interface quality matters. It should define:

  • purpose of the UI;
  • product feel and design principles;
  • primary screens/surfaces;
  • screen-by-screen notes;
  • layout and information architecture;
  • component inventory;
  • key user flows;
  • states to design: empty, loading, error, success, first-run, permissions;
  • visual direction;
  • accessibility and responsiveness;
  • content/copy notes;
  • optional image-generation concepts;
  • selected direction;
  • handoff notes for implementation.
Optional image-generation pass

Image generation is concept exploration, not the implementation source of truth.

Use it when:

  • the user wants possible UI designs;
  • taste is ambiguous;
  • a dashboard/app/landing page needs visual exploration before code;
  • comparing 2-3 directions would help.

Default concepts:

  1. Conservative — familiar, low-risk, easy to implement.
  2. Strong-fit — best interpretation of the product brief.
  3. Divergent — more distinctive, useful for taste discovery.

If image generation is unavailable or skipped, write the prompts and continue text-only. If images are generated, use vision/manual review to translate them back into written requirements, component notes, and acceptance criteria. Do not ask Superpowers to copy an image blindly.

Stage 5: Research Pass

For the specific idea, look for similar products and note:

  • what already exists;
  • what is common / commodity;
  • what is different;
  • what should be avoided;
  • where the idea fits.

Keep this practical and bounded. Research should strengthen the spec, not become an endless market report.

Stage 6: Implementation Thinking

Translate the idea into 03-implementation-spec.md if the UI brief exists, otherwise 02-implementation-spec.md.

Include:

  • major system pieces;
  • data needs;
  • recommended database/storage choice and why;
  • hosting/deployment target;
  • platform target decisions;
  • app topology decisions;
  • frontend structure and how it respects the UI design brief if present;
  • backend/service needs;
  • recommended technical stack defaults;
  • security/secrets model;
  • integration points;
  • workflow/milestones;
  • risks and tradeoffs;
  • executable build tasks;
  • testing and verification plan;
  • acceptance criteria / done means;
  • a build sequence a developer or AI coding agent could follow.

Use recommend-then-confirm for technical choices:

  1. infer sensible defaults from product constraints;
  2. present a concise recommendation table;
  3. ask whether the user accepts or changes items;
  4. record accepted defaults and explicit overrides;
  5. if unsure, proceed with defaults as assumptions.

Stage 7: Final Agent Build Handoff

Create or update README.md using templates/idea-package-readme-template.md.

Create the final handoff as:

  • 04-agent-build-handoff.md if the UI brief exists;
  • 03-agent-build-handoff.md if the UI brief was skipped.

Load idea-to-implementation-doc when creating the handoff because the canonical handoff template lives there.

The handoff must include:

  • mission;
  • product vision;
  • non-negotiable requirements;
  • UI/design direction or explicit UI assumptions;
  • out of scope;
  • technical architecture;
  • implementation phases;
  • build tasks;
  • testing requirements;
  • verification commands/checks;
  • acceptance criteria;
  • done means checklist;
  • prompt for the build agent;
  • explicit Superpowers handoff instructions.

Stage 8: Spec Review / Readiness Gate

Create:

  • 05-spec-review.md if the UI brief exists;
  • 04-spec-review.md if the UI brief was skipped.

Review whether:

  • the product goal is clear;
  • UI/design expectations are clear enough for a fresh agent to avoid generic output;
  • requirements are testable;
  • unresolved product decisions remain;
  • unresolved technical decisions remain;
  • acceptance criteria are concrete;
  • done means is specific;
  • a fresh agent could build from the file without asking obvious questions;
  • testing and verification requirements are included;
  • non-goals prevent scope creep.

Verdicts:

  • PASS — ready to feed into Superpowers.
  • PASS WITH CHANGES — mostly ready; patch listed issues first.
  • FAIL — too ambiguous or incomplete.

Stage 9: Review / Plan Review

If the user wants a critique pass, review artifacts separately:

  • design doc stays product-focused;
  • UI brief gives concrete screen/style/state guidance without prescribing unnecessary code;
  • implementation doc stays build-focused;
  • final handoff is agent-ready;
  • spec review is honest about gaps;
  • package summary does not duplicate all focused docs.

Progression Rules and Override Phrase

Lite mode rules

  • Capture quickly; do not force research, UI brief, or implementation planning.
  • Ask at most 3-5 clarifying questions before drafting unless the user wants more.
  • Do not create 02-ui-design-brief.md, implementation spec, build handoff, or spec review unless the user upgrades to Full mode or specifically asks for UI/mockup direction.
  • If the user says the idea is just a note, keep it as a note.

Full mode rules

  • Do not move from capture to design until target user, problem, and core behavior are clear enough to summarize.
  • Do not move from design to implementation if open product or UI questions would change architecture, MVP scope, or core interface.
  • Do not create final build handoff until implementation phases, testing requirements, acceptance criteria, and done means are present.
  • Do not mark spec review PASS if verification checks are missing, major product/technical decisions remain, or UI expectations are too vague for an interface-heavy app.
  • If the UI stage is skipped, carry UI assumptions forward instead of blocking.

User override phrase

The user can force progression with this exact phrase:

GREENLIGHT NEXT STAGE

When the user says it, move to the next stage. Briefly note the risk, carry unresolved items under Open Questions / Assumptions, and continue. This does not allow unsafe actions or credential exposure.

Operating Rules

  • Keep the workflow modular.
  • Prefer separate files for separate stages.
  • Save progress as Markdown.
  • Patch the working idea note when the user confirms product, architecture, or UI decisions.
  • Keep reusable skill content public-safe; do not embed private product plans or credentials.
  • If the user says stop, that's enough, enough for now, done, draft it, or write it up, stop questioning and draft immediately.
  • If the user wants a lighter experience, use Lite mode.
  • If a stage is skipped, continue with the previous workflow and record assumptions.

Relationship to Focused Skills

  • Use idea-to-design-doc when the user only wants the design stage.
  • Use idea-to-ui-design-brief when the user wants UI direction, generated UI concepts, screen-by-screen mockup notes, or post-build UI redesign.
  • Use idea-to-implementation-doc when the user only wants implementation planning.
  • Use idea-superpowers-suite when the user wants the full system.

Relationship to Superpowers

The idea workflow is the front-end product/spec pipeline. Superpowers is the execution discipline pipeline.

Expected handoff:

  1. idea-superpowers-suite captures the idea and asks clarifying questions.
  2. idea-to-design-doc produces the design/spec document.
  3. idea-to-ui-design-brief optionally produces UI direction and image-generation concept prompts.
  4. idea-to-implementation-doc produces the build-ready Markdown handoff.
  5. Superpowers validates, plans, implements, tests, reviews, and verifies.

The idea workflow should create high-quality input for Superpowers, not replace Superpowers execution.

Reference Files

  • references/interview-question-bank.md — reusable question bank for Lite and Full interviews.
  • references/example-cli-tool-build-handoff.md — generic example final handoff for a small developer CLI tool.
  • references/example-saas-web-app-build-handoff.md — generic example final handoff for a small SaaS/web app.
  • references/example-automation-script-build-handoff.md — generic example final handoff for a practical automation workflow.

Templates

  • idea-to-implementation-doc/templates/agent-build-handoff-template.md — required structure for final agent/Superpowers build handoff documents.
  • templates/idea-package-readme-template.md — README/status index for Full mode idea folders.

Common Pitfalls

  1. Making the UI stage mandatory. It is optional and Full-mode only by default.
  2. Letting generated images become the source of truth. Translate images into written UI requirements.
  3. Renumbering existing packages inconsistently. Preserve established numbering and update README maps.
  4. Overloading Lite mode. Keep Lite fast unless the user asks for UI or Full mode.
  5. Producing generic UI slop. Use product-specific screens, states, and actions.
  6. Moving into code before product and UI decisions that affect scope are clear.

Verification Checklist

  • Mode was chosen or inferred correctly.
  • Lite mode stayed lightweight unless explicitly upgraded.
  • Full mode artifacts are separated by stage.
  • UI brief exists only when useful/requested, or UI assumptions are carried forward.
  • Optional image-generation prompts are treated as concepts, not implementation truth.
  • README artifact map matches actual filenames and numbering.
  • Final handoff includes UI direction or explicit UI assumptions.
  • Spec review honestly reports unresolved product, UI, and technical gaps.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Idea Superpowers Suite AI skill do?

Use when running the full idea workflow: capture a rough idea, expand it into design/UI/implementation docs, research similar products, and generate build-ready Markdown artifacts.

Why use Idea Superpowers Suite on TypingMind?

Because you install it once and use it with any model. Idea Superpowers Suite is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Idea Superpowers Suite in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/AkoliteZA/hermes-agent-idea-workflow/tree/main/idea-superpowers-suite. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Idea Superpowers Suite?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Idea Superpowers Suite?

As many as you like. As long as a model supports skills, you can use Idea Superpowers Suite with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Idea Superpowers Suite AI skill free?

Yes. It is published on GitHub by AkoliteZA under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇