Skills2026年4月9日·1 分钟阅读

Awesome Claude Code Subagents — 130+ Specialized Agents

Install 130+ specialized Claude Code subagents across 10 categories: core dev, language experts, infra, security, data/AI, DevEx, and business. Plugin-based with 16.7K GitHub stars.

Agent 就绪

先审查再安装

这个资产需要先审查。复制的指令会要求 Agent dry-run、列出写入项,确认后再继续。

Needs Confirmation · 64/100策略:需确认
Agent 入口
任意 MCP/CLI Agent
类型
Skill
安装
Single
信任
信任等级:Established
入口
Awesome Claude Code Subagents
先审查命令
npx -y tokrepo@latest install 0929e53d-33aa-11f1-9bc6-00163e2b0d79 --target codex

先 dry-run,确认写入项后再运行此命令。

TL;DR
A curated TokRepo workflow guide for Awesome Claude Code Subagents.
§01

What it is

Awesome Claude Code Subagents — 130+ Specialized Agents is a public TokRepo workflow curated around the upstream project at VoltAgent/awesome-claude-code-subagents.

It is best for developers who want a repeatable, copy-pasteable setup that starts from the workflow steps (not marketing claims) and links back to the canonical upstream docs.

Quick facts (verified sources):

  • GitHub stars: 20219
  • Last pushed: 2026-05-20T19:25:48Z
  • License (SPDX): MIT
  • TokRepo view_count: 249

From upstream README (for context):

<a href="https://github.com/VoltAgent/voltagent">

<img width="1500" height="500" alt="Group 32" src="https://github.com/user-attachments/assets/55b97c47-8506-4be0-b18f-f5384d063cbb" />

</a>

<br />

<br/>

<div align="center">

§02

How it saves time or tokens

This workflow saves time by packaging a “known-good starting path” into a single, reusable page: you get the upstream repo link, the workflow’s step-by-step instructions, and a short set of pitfalls to avoid.

If you run agents or CLI tools repeatedly, the biggest cost is usually re-discovering the same setup details and re-checking prerequisites. A curated workflow reduces that repeated context-building and keeps your prompts shorter because you can point your agent back to a stable set of steps and citations.

§03

How to use

  1. Awesome Claude Code Subagents
§04

Example

§05

Quick Use\n\nInstall all subagents via the plugin marketplace:\n``bash\n/plugin marketplace add https://github.com/VoltAgent/awesome-claude-code-subagents\n/plugin install voltagent-core-dev\n/plugin install voltagent-lang\n/plugin install voltagent-infra\n`\n\nOr install the standalone script:\n`bash\ncurl -fsSL https://raw.githubusercontent.com/VoltAgent/awesome-claude-code-subagents/main/install.sh | bash\n`\n\n---\n\n## Intro\n\nAwesome Claude Code Subagents is a collection of 130+ specialized AI assistants organized across 10 categories, with 16,700+ GitHub stars. Each subagent is a domain expert — from TypeScript pro to Kubernetes specialist to product manager — that you install as a Claude Code plugin. One command adds an entire category of expertise to your development workflow. Setup time: under 5 minutes.\n\nBest for: development teams needing specialized AI expertise across the full stack. Works with: Claude Code plugins, compatible with Cursor and Codex via manual install.\n\n---\n\n## 130+ Subagents Organized in 10 Categories\n\n### 01. Core Development (voltagent-core-dev)\napi-designer, backend-developer, design-bridge, electron-pro, frontend-developer, fullstack-developer, graphql-architect, microservices-architect, mobile-developer, ui-designer, websocket-engineer — 11 agents\n\n### 02. Language Specialists (voltagent-lang)\ntypescript-pro, python-pro, golang-pro, rust-engineer, java-architect, react-specialist, nextjs-developer, vue-expert, angular-architect, django-developer, fastapi-developer, flutter-expert, swift-expert, kotlin-specialist, rails-expert, laravel-specialist, spring-boot-engineer, and more — 29 agents covering every major tech stack\n\n### 03. Infrastructure (voltagent-infra)\ncloud-architect, docker-expert, kubernetes-specialist, terraform-engineer, devops-engineer, sre-engineer, platform-engineer, database-administrator, network-engineer, deployment-engineer, incident-responder — 16 agents\n\n### 04. Quality & Security (voltagent-qa-sec)\ncode-reviewer, security-auditor, penetration-tester, qa-expert, test-automator, debugger, performance-engineer, compliance-auditor, accessibility-tester, chaos-engineer — 15 agents\n\n### 05. Data & AI (voltagent-data-ai)\nai-engineer, data-scientist, ml-engineer, llm-architect, data-analyst, data-engineer, prompt-engineer, nlp-engineer, mlops-engineer, postgres-pro, database-optimizer — 13 agents\n\n### 06. Developer Experience (voltagent-dev-exp)\ndocumentation-engineer, cli-developer, refactoring-specialist, mcp-developer, build-engineer, git-workflow-manager, legacy-modernizer, readme-generator, tooling-engineer, dx-optimizer — 14 agents\n\n### 07. Specialized Domains (voltagent-domains)\nblockchain-developer, fintech-engineer, game-developer, iot-engineer, embedded-systems, payment-integration, seo-specialist, api-documenter — 12 agents\n\n### 08. Business & Product (voltagent-biz)\nproduct-manager, project-manager, technical-writer, business-analyst, ux-researcher, scrum-master, content-marketer, sales-engineer, legal-advisor — 12 agents\n\n### 09. Meta & Orchestration (voltagent-meta)\nmulti-agent-coordinator, workflow-orchestrator, task-distributor, context-manager, knowledge-synthesizer, agent-organizer, performance-monitor — 13 agents\n\n### 10. Research & Analysis (voltagent-research)\nresearch-analyst, search-specialist, trend-analyst, competitive-analyst — 4 agents\n\n### Installation Methods\n\n| Method | Command |\n|--------|---------|\n| Plugin Marketplace | /plugin marketplace add [url] |\n| Manual | Copy agent files to .claude/plugins/ |\n| Interactive Installer | Run install.sh and select categories |\n| Standalone Script | curl -fsSL ... \\| bash |\n| Agent Installer | Ask Claude to install specific agents |\n\n### FAQ\n\nQ: What are Claude Code Subagents?\nA: Specialized AI assistants that you install as Claude Code plugins, each with domain-specific expertise for tasks like code review, DevOps, data science, or product management.\n\nQ: Is Awesome Claude Code Subagents free?\nA: Yes, the collection is open source and free to install.\n\nQ: Can I install only specific categories?\nA: Yes, each category is a separate plugin (e.g., voltagent-core-dev, voltagent-lang). Install only what you need.\n\n---\n\n## Source & Thanks\n\n> Created by VoltAgent.\n>\n> awesome-claude-code-subagents — ⭐ 16,700+\n\nThanks to VoltAgent for building the most comprehensive subagent collection for Claude Code.\n\n---\n\n<!-- ZH -->\n\n## Quick Use\n\nInstall all subagents via the plugin marketplace:\n`bash\n/plugin marketplace add https://github.com/VoltAgent/awesome-claude-code-subagents\n/plugin install voltagent-core-dev\n/plugin install voltagent-lang\n/plugin install voltagent-infra\n`\n\nOr use the one-click install script:\n`bash\ncurl -fsSL https://raw.githubusercontent.com/VoltAgent/awesome-claude-code-subagents/main/install.sh | bash\n``\n\n---\n\n## Introduction\n\nAwesome Claude Code Subagents is a collection of 130+ specialized AI assistants organized into 10 categories, with 16,700+ GitHub stars. Each subagent is a domain expert — from TypeScript master to Kubernetes expert to product manager — installable as Claude Code plugins. One command adds an entire category of specialized capabilities.\n\n---\n\n## 10 Categories, 130+ Specialized Subagents\n\nCore Development (11): API designer, frontend/backend/full-stack developer, microservice architect, and more\nLanguage Experts (29): Full-stack coverage across TypeScript, Python, Go, Rust, React, Next.js, Vue, and more\nInfrastructure (16): Cloud architecture, Docker, K8s, Terraform, DevOps, SRE, and more\nQuality & Security (15): Code review, security audit, penetration testing, QA, performance engineering, and more\nData & AI (13): AI engineer, data scientist, ML engineer, LLM architect, and more\nDeveloper Experience (14): Documentation engineering, CLI development, refactoring specialist, MCP development, and more\nSpecialized Domains (12): Blockchain, fintech, game development, IoT, and more\nBusiness & Product (12): Product manager, project manager, tech writer, UX researcher, and more\nOrchestration (13): Multi-agent coordinators, workflow orchestration, task dispatch, and more\nResearch & Analysis (4): Research analyst, trend analyst, and more\n\n---\n\n## Source & Thanks\n\n> Created by VoltAgent.\n>\n> awesome-claude-code-subagents — ⭐ 16,700+\n

§06

Related on TokRepo

§07

Common pitfalls

  • Skipping the upstream README and relying on a copied snippet without checking prerequisites (OS, runtime, permissions).
  • Treating example configs as production-ready without reviewing secrets handling and access control.
  • Not pinning versions (CLI/tools) and then debugging breakages after automatic upgrades.

Operational checklist (generic, verify against upstream docs)

  • Confirm prerequisites (runtime version, OS support, system packages).
  • Keep secrets out of the repo (env vars or a secret manager).
  • Start with the smallest end-to-end action and expand only after it works.
  • Add timeouts, retries, and clear logs before you run this in CI.
  • Record the exact versions you tested (tool, runtime, dependencies).

How to adapt this workflow for a team

If more than one person will run this, treat the workflow like a small runbook. Write down: (1) the baseline command that proves it works, (2) where credentials live, and (3) what “good output” looks like. Then make changes one at a time: pin versions, add a wrapper script, and only then integrate into automation. This keeps troubleshooting simple because you always have a known-good reference path to compare against.

Security and reliability notes (generic)

Before you automate, do a quick threat-model pass: what data flows into the tool, what leaves it, and what gets stored. Avoid pasting secrets into prompts or config committed to git. If the workflow calls remote services, document rate limits and error handling; transient failures are normal, so your automation should degrade gracefully. If you store artifacts (logs, caches, indexes), decide retention and access control up front.

常见问题

What is Awesome Claude Code Subagents?+

Awesome Claude Code Subagents is a TokRepo workflow page that curates a specific upstream GitHub project and the exact steps needed to start using it. Instead of relying on unverified platform claims, the workflow is designed to be a repeatable setup path: follow the workflow steps, cross-check any prerequisites against the upstream README, and keep the repository as the source of truth. This is most useful when you reuse the same tool across multiple projects and want the setup to stay consistent over time.

What do I need before running this workflow?+

Start by reading the upstream README and comparing it with the TokRepo workflow steps. Common prerequisites include a supported runtime (Node/Python/Go), OS-specific dependencies, and required credentials or environment variables. If the workflow uses a CLI or a server, record the exact version you install so teammates can reproduce your environment. When in doubt, run the smallest possible command first and only then expand to more advanced configuration, so failures are easy to isolate.

How do I validate it end-to-end after setup?+

Use an end-to-end smoke test that matches the workflow’s goal. For a CLI, that might be a single version/help command followed by one minimal action. For an MCP integration, start with tool discovery (list/describe tools) before calling any tool, so you confirm the client-server contract is working. For a server, verify a health endpoint or a trivial request first. Keep the exact command lines and logs you used; they are the fastest debug path when upstream behavior changes.

Is it free to use, and what license applies?+

License terms come from the upstream repository, not TokRepo. This workflow includes a citation to the upstream LICENSE so you can verify usage and redistribution rights for your scenario. GitHub metadata reports the SPDX identifier as MIT, but treat the LICENSE file itself as authoritative because repositories can include exceptions or multiple license files. If you plan to bundle or redistribute, do a quick license check before you automate the workflow.

What are common pitfalls when using workflows like this?+

The most common pitfall is copying a snippet without verifying prerequisites and then debugging environment issues that are documented upstream. The next pitfall is secrets handling: example configs often contain placeholders, and teams accidentally commit real tokens. Finally, workflows can drift when upstream changes (new releases, changed defaults). Pin versions where possible, and re-check upstream docs periodically; the repository’s activity timestamp (2026-05-20T19:25:48Z) is a useful signal for how frequently you should expect change.

引用来源 (3)
🙏

来源与感谢

Created by VoltAgent.

awesome-claude-code-subagents — ⭐ 16,700+

讨论

登录后参与讨论。
还没有评论,来写第一条吧。

相关资产