Skill Development Limited info

SuperClaude_Framework

A meta-programming framework that enhances Claude Code with 30 slash commands, 20 specialized agents, 7 behavioral modes, and 8 MCP server integrations.

  • 23.9k GitHub stars
View on GitHub (opens in a new tab) Setup guide
pipx install superclaude
SuperClaude_Framework — illustration

What it is

SuperClaude Framework is a meta-programming configuration that transforms Claude Code into a structured development platform. It works through behavioral instruction injection and component orchestration, adding 30 slash commands, 20 specialized agents, 7 behavioral modes, and 8 MCP server integrations. The framework is installed via pipx or a direct git clone and enhances the existing Claude Code session with additional capabilities. The 30 commands cover the complete development lifecycle: /sc:research for deep web research, /sc:brainstorm for structured ideation, /sc:implement for code implementation, /sc:test for testing workflows, and /sc:pm for project management, among others. The 20 specialized agents include PM Agent, Deep Research Agent, Security Engineer, Frontend Architect, and domain-specific experts. These agents coordinate automatically based on the context of the task rather than requiring manual selection. Seven behavioral modes adapt the framework's behavior: brainstorming mode for strategic questioning, business panel mode for multi-expert analysis, deep research mode for autonomous investigation, token-efficiency mode for context optimization, plus orchestration, task management, and introspection modes. The deep research capability (v4.2+) features adaptive planning with three strategies, multi-hop reasoning with up to 5 iterative searches, and confidence-based quality scoring on a 0.0 to 1.0 scale.

Who it's for

  • Claude Code power users
  • software developers
  • engineering teams
  • AI workflow enthusiasts

Requirements

Requirements

  • Claude Code
  • Python with pipx (recommended) or Git

Setup

  1. Step 1

    bash
    pipx install superclaude

  2. Step 2

    bash
    superclaude install

إطار برمجة وصفية يعزز Claude Code بـ 30 أمرا مختصرا و 20 وكيلا متخصصا و 7 أوضاع سلوكية و 8 خوادم MCP.

Examples

Deep research

Prompt
prompt
/sc:research [topic]

Expected output: Triggers the deep research agent with adaptive planning, multi-hop reasoning (up to 5 iterative searches), and confidence-based quality scoring.

Code implementation

Prompt
prompt
/sc:implement [feature description]

Expected output: Invokes the implementation workflow with automatic agent coordination based on context.

Pros & cons

Pros

  • Pro:30 commands covering the complete development lifecycle
  • Pro:20 specialized agents that coordinate automatically based on context
  • Pro:7 behavioral modes from brainstorming to token-efficiency
  • Pro:Multi-hop deep research with confidence scoring (0.0-1.0 scale)

Cons

  • Con:Not affiliated with Anthropic; community-maintained
  • Con:Requires Claude Max ($100/month) for full testing capacity