1. Conduid
  2. Commerce
  3. Super Powers
MCP server · Commerce

Super Powers

Repository created from TEMPLATES main branch

Unclaimed commerce
34Low

Scored 6 months ago · breakdown

About Super Powers

Super Powers is an MCP server in the Commerce category: repository created from TEMPLATES main branch.

Install

From source
git clone https://github.com/paulanunes85/super-powers

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Super Powers

Powered by Claude · Grounded in docs

I know everything about Super Powers. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

  • ✓README presentThe repository ships documentation.

5 checks have not been run yet.

README

TEMPLATES

Use VS Code + GitHub Copilot as a productivity platform for any professional — create documents, reports, dashboards, presentations, diagrams, spreadsheets, and everything else you need in your daily work.

No code. No terminal. Just prompts, agents, and skills working for you.

Clone this repo, run /onboard, and start generating enterprise-grade deliverables in minutes.

This is a GitHub Template. Click "Use this template" on the repo page to create your own copy.

Quick Start

# 1. Use this template (click "Use this template" on GitHub) or clone
git clone https://github.com/<your-org>/TEMPLATES.git
cd TEMPLATES

# 2. Start the MCP server (Docker required)
cd mcp-servers
cp .env.example .env
make up

# 3. Open in VS Code
cd .. && code .

# 4. Run the onboarding wizard
# In VS Code Copilot chat, type: /onboard

Post-Clone Checklist

After using this template, update these files with your info:

File What to Change
context/CLIENT.md Run /onboard to fill automatically
logos/customer/ Add your logo files (color + white variants)
LICENSE Replace <Your Name or Organization>
README.md Update clone URL with your repo

What You Can Create

Deliverable Command Agent
Word document /docx DOCX Creator
PowerPoint deck /pptx PPTX Creator
Gamma presentation /gamma Gamma Presenter
Excel dashboard /xlsx XLSX Creator
PDF report /pdf PDF Creator
Architecture diagram /html-diagram HTML Diagrammer
FigJam visual /figma-diagram Figma Diagrammer
Markdown doc /markdown Markdown Writer
Meeting summary /process-transcription Meeting Analyst
Weekly sales report /weekly-sales-status Markdown Writer
Interactive website (ask Site Builder) Site Builder

Every output is saved to output/ with versioning, archiving, and Microsoft branding applied automatically.

What's Inside

15 Agent Modes

Custom persistent modes you can switch to in VS Code or Claude Code.

Agent Purpose
Orchestrator Interactive onboarding — sets up a new workspace with structured questions
DOCX Creator Microsoft-branded Word documents (.docx) with 4-color palette
PPTX Creator PowerPoint presentations with 10-slide structure and speaker notes
Gamma Presenter Gamma presentations with Microsoft theme and executive formatting
Figma Diagrammer FigJam diagrams via Mermaid syntax with rendering-safe patterns
HTML Diagrammer Interactive React-based HTML diagrams with language selector (EN/PT-BR/ES)
XLSX Creator Enterprise Excel workbooks with dashboards, charts, and connected tabs
PDF Creator Microsoft-branded PDF documents with cover pages, KPI cards, and tables
Markdown Writer Versioned markdown documents with YAML frontmatter and ToC
Site Builder Multi-page interactive websites from project documentation
Meeting Analyst Process Teams .vtt transcriptions into structured summaries with insights
Agent Creator Scaffold new agents, skills, prompts, and instructions
Skill Architect Expert skill designer — structures domain knowledge with quality checklists
Agent Validator Audit AI customization files for correctness and discoverability
Prompt Engineer Design production-ready prompts for Claude models

16 Slash Commands (Prompts)

One-shot tasks you trigger with /command in the chat.

Command What It Does
/onboard Interactive workspace setup wizard
/docx Generate a Microsoft-branded Word document
/pptx Generate a 10-slide PowerPoint presentation
/gamma Generate a Gamma presentation
/figma-diagram Create a FigJam diagram with Mermaid syntax
/html-diagram Create an interactive HTML architecture diagram
/xlsx Generate an Excel workbook with dashboard and charts
/pdf Generate a Microsoft-branded PDF document
/markdown Generate a versioned Markdown document
/weekly-sales-status Generate a weekly sales pipeline status report
/process-transcription Parse a Teams .vtt file into a meeting summary
/review Run a security + quality + performance code review
/spec-driven Run the Spec-Driven Development workflow (5 phases)
/agents-md Scaffold an AGENTS.md file for any project
/create-skill Scaffold a new SKILL.md with proper structure
/create-mcp-server Scaffold a complete MCP server project (TS/Python/C#)

10 Skills (Domain Knowledge)

Skills are loaded by agents before generating output — they contain branding, templates, and quality checklists.

Skill Domain
docx-creator Word document branding, structure, and formatting
pptx-creator PowerPoint 10-slide structure and design rules
ms-gamma-presenter Gamma theme, slide templates, and visual guide
figjam-diagrams Mermaid rendering rules, color palette, proven examples
html-diagrams React HTML template, language selector, interactive patterns
markdown-writer Frontmatter, versioning, 7 document type templates
xlsx-creator Excel workbook branding, dashboard layout, chart patterns
pdf-creator PDF branding, cover page, table formatting, KPI cards
weekly-sales-status Weekly pipeline report structure and sales cadence
react-best-practices 40+ React/Next.js performance optimization rules
web-design-guidelines Web Interface Guidelines, accessibility, WCAG compliance
mcp-ecosystem 39 MCP tools reference from 9 open-source projects

5 Auto-Applied Instructions

Rules that activate automatically when editing specific file types.

Instruction Applies To Key Rules
agent-files *.agent.md, *.prompt.md, SKILL.md Frontmatter standards, description formula, DO/DON'T examples
typescript *.ts ES modules, strict types, async patterns, naming conventions
dockerfile Dockerfile Multi-stage builds, security, layer optimization
docker-compose docker-compose.yml Health checks, localhost binding, named volumes
mermaid-figjam *.mmd, *.mermaid FigJam rendering safety, Microsoft brand colors

Architecture

TEMPLATES/
├── .github/
│   ├── agents/           # 15 custom agent modes (.agent.md)
│   ├── instructions/     # 5 auto-applied rules (.instructions.md)
│   ├── prompts/          # 16 slash commands (.prompt.md)
│   ├── skills/           # 10 domain knowledge packs (SKILL.md)
│   └── copilot-instructions.md  # Workspace-wide rules (always loaded)
│
├── context/              # Working context (gitignored)
│   ├── CLIENT.md         # Master profile (template)
│   ├── transcriptions/   # Teams .vtt recordings
│   ├── emails/           # Email threads as markdown
│   ├── messages/         # Teams/Slack messages
│   ├── documents/        # Shared documents
│   ├── notes/            # Manual observations
│   └── journal/          # Meeting summaries and decisions
│
├── output/               # Generated deliverables (gitignored)
│   ├── docx/             # Word documents + archive/
│   ├── pptx/             # PowerPoint files + archive/
│   ├── html/             # HTML diagrams + archive/
│   ├── pdf/              # PDF documents + archive/
│   ├── xlsx/             # Excel workbooks + archive/
│   └── md/               # Markdown documents + archive/
│
├── mcp-servers/          # Local MCP ecosystem (Docker)
│   ├── src/tools/        # 9 tool modules (39 tools total)
│   ├── docker-compose.yml
│   └── Makefile
│
├── docs/                 # Reference guides and best practices
├── logos/                # Logo assets for documents and presentations
├── sources/              # Reference materials (PDFs, guides)
├── AGENTS.md             # For AI coding agents (Copilot, Claude Code, Codex)
├── CLAUDE.md             # For Claude Code sessions
├── CONTRIBUTING.md       # How to add agents, skills, prompts, instructions
└── README.md             # This file

AI Primitives Hierarchy

This repo uses 5 types of AI customization files that work together:

┌─────────────────────────────────────────────────┐
│  copilot-instructions.md    (always loaded)      │
│  ┌───────────────────────────────────────────┐  │
│  │  .instructions.md   (auto-apply by file)  │  │
│  │  ┌───────────────────────────────────┐    │  │
│  │  │  .agent.md  ←loads→  SKILL.md     │    │  │
│  │  │       ↑                           │    │  │
│  │  │  .prompt.md  (routes via agent:)  │    │  │
│  │  └───────────────────────────────────┘    │  │
│  └───────────────────────────────────────────┘  │
└─────────────────────────────────────────────────┘
  • Agents load skills via Step 0: Load Skill Knowledge before generating
  • Prompts route to agents via the agent field in frontmatter
  • Instructions auto-apply when editing matching file types
  • Skills are the single source of truth — agents never hardcode what skills define

MCP Server

A local MCP ecosystem server running at http://localhost:3100/mcp via Docker. Exposes 39 tools from 9 open-source projects:

Source Tools What It Provides
spec-kit 5 Spec-Driven Development methodology
anthropics/skills 5 Anthropic skill catalog and templates
awesome-copilot 4 Community agents, skills, and prompts
agent-framework 4 Microsoft agent architecture patterns
gh-aw 4 GitHub Agentic Workflows
agents.md 3 AGENTS.md format specification
Claude Code 4 Claude Code documentation
GitHub Copilot Docs 5 GitHub Copilot documentation — customization, extensions, features
Anthropic Platform Docs 5 Anthropic API, Agent SDK, models, prompt engineering, skills
cd mcp-servers && make up       # Start (auto-restarts on boot)
cd mcp-servers && make health   # Verify: {"status":"ok"}
cd mcp-servers && make logs     # Tail logs
cd mcp-servers && make down     # Stop

Client Workflow

1. Set up your workspace

Run /onboard in VS Code Copilot chat. The Orchestrator agent will:

  • Ask structured questions (your info, strategic themes, context)
  • Fill context/CLIENT.md with real data
  • Start the MCP server
  • Validate the workspace is ready
  • Suggest first actions

2. Create deliverables

Need Command Output
Executive presentation /gamma or /pptx 10-slide deck with Microsoft branding
PDF document /pdf Branded PDF with cover page and KPI cards
Excel dashboard /xlsx Workbook with KPIs, charts, connected tabs
Technical document /docx Word document with 4-color palette
Architecture diagram /html-diagram Interactive HTML with click-to-explore
FigJam visual /figma-diagram Mermaid diagram in FigJam
Meeting notes /process-transcription Structured summary from .vtt
Weekly sales report /weekly-sales-status Pipeline status with committed, uncommitted, PBO
Technical spec /markdown Versioned markdown with frontmatter
Code review /review Security + quality + performance audit
MCP server scaffold /create-mcp-server Complete MCP project (TS/Python/C#)

3. Build knowledge over time

Every meeting transcription, email, and document feeds into context/. The Meeting Analyst agent processes these into structured journal entries, keeping your workspace context rich and organized.

Branding

All generated documents use Microsoft branding by default:

Element Value
4-color palette Red #F25022, Green #7FBA00, Blue #00A4EF, Yellow #FFB900
Primary blue #0078D4
Extended Green #107C10, Red #E81123, Teal #008272, Purple #5C2D91, Orange #D83B01
Font Segoe UI (body), Segoe UI Semibold (headings)
Text #323130 (primary), #605E5C (secondary)

Factual Integrity

All agents follow strict factual integrity rules:

  • NEVER fabricate metrics, KPIs, ROI, market data, or statistics
  • Every data claim must include a hyperlink to a credible source
  • Credible sources: Gartner, Forrester, Microsoft Learn, GitHub Docs, Anthropic Blog, Azure Architecture Center
  • If no source exists, state as assumption or omit entirely
  • Every document includes a References section

Document Versioning

All generated documents follow this pattern:

  • Filename: {Title}_v{version}_{YYYY-MM-DD}.{ext}
  • Save to: output/{format}/ (e.g., output/docx/)
  • Archive: Previous versions move to output/{format}/archive/
  • Metadata: Every document includes date, version (semver), and author name

Setup for Different Clients

VS Code (GitHub Copilot)

The .vscode/settings.json auto-registers the MCP server. Just open the folder in VS Code.

Claude Code

claude mcp add mcp-ecosystem --transport http --url http://localhost:3100/mcp

The .claude/settings.json pre-configures permissions.

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "mcp-ecosystem": {
      "url": "http://localhost:3100/mcp"
    }
  }
}

Security

  • Context data (context/) is gitignored — never committed
  • Generated output files are gitignored — only .gitkeep structure is tracked
  • MCP server binds to localhost only
  • Docker containers run as non-root user
  • .env files are gitignored — use .env.example as reference
  • No API keys or tokens in the repository

Contributing

See CONTRIBUTING.md for how to add new agents, skills, prompts, and instructions.

License

MIT License. See LICENSE for details.

README copy captured when this server was indexed; it may lag the repository. The original is authoritative.

Questions

About Super Powers

How do I install Super Powers?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Super Powers safe to use with an AI agent?

Its trust score is 34 out of 100 (low). It passes 1 of the 1 static security checks run so far. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Super Powers still maintained?

Conduid hasn't recorded a commit date for this repository yet. Check the repository directly for recent activity.