Report Abuse

Basic Information

This repository is a curated collection of 75 specialized subagents for Claude Code designed to extend Claude Code's development and operational workflows with domain-specific expertise. Each subagent is defined by a name, description, optional model assignment, and a system prompt so Claude Code can automatically delegate tasks or invoke a named subagent on explicit request. The collection spans development and architecture, language specialists, infrastructure and operations, quality and security, data and AI, documentation, SEO and marketing, and several specialized domains. Subagents are assigned to one of three models based on complexity to balance capability and cost. The README documents installation into the ~/.claude/agents directory, usage examples, multi-agent orchestration patterns, and contribution guidelines for adding new subagents.

Links

Categorization

App Details

Features
A library of 75 ready-to-use subagents organized by domain with explicit subagent format and examples. Per-subagent model configuration allows selection among three Claude models (haiku, sonnet, opus) to match task complexity and cost. Automatic invocation by Claude Code based on context plus explicit invocation by name. Multi-agent orchestration patterns and example workflows including sequential, parallel, conditional branching, and review/validation steps. Companion slash-commands collection for advanced workflows and prebuilt orchestration. Installation instructions for drop-in use under ~/.claude/agents and contribution guidance for adding new .md subagent files. Usage examples cover single-agent tasks, multi-agent workflows, and advanced orchestration patterns. MIT licensed.
Use Cases
The collection accelerates developer and operations work by providing domain experts for common tasks such as code review, security audits, architecture design, performance optimization, database tuning, CI/CD configuration, data analysis, RAG and LLM engineering, and documentation generation. It standardizes how capabilities are described and invoked so teams can rely on consistent, testable agent behaviors. Model assignments let teams reduce cost by using smaller models for simple work and higher-capability models for critical tasks. Prebuilt workflows and examples demonstrate how to chain specialists for feature development, incident response, ML pipelines, and product launches. Easy installation and extension let organizations adopt and modify subagents to fit their stacks and processes.

Please fill the required fields*