</>MCP Agents Market
Agent

awesome-codex-subagents

by VoltAgent6kUpdated 2026-07-10

A collection of 130+ specialized Codex subagents covering a wide range of development use cases.

Codex

Awesome Codex Subagents is a comprehensive repository of 171+ specialized AI subagent definitions organized across 13 categories, designed specifically for the Codex AI coding assistant. Each subagent is packaged as a .toml configuration file containing role-specific instructions, model routing preferences, and sandbox permissions that developers copy into their Codex agent directories (~/.codex/agents/ for global use or .codex/agents/ for project-specific use). These subagents provide task-specific expertise ranging from backend development and security auditing to data science and platform engineering, with explicit delegation workflows that maintain isolated context windows for each specialized task. The collection enables development teams to build consistent, reusable AI workflows across projects by sharing standardized subagent definitions.

Key Features

171+ pre-configured subagent definitions across 13 specialized categories including Core Development, Language Specialists, Infrastructure, Quality & Security, and Data & AI
Codex-native .toml format with structured instructions, smart model routing (GPT-5.4 for deep reasoning, GPT-5.3-codex-spark for fast tasks), and configurable sandbox modes
Dual installation paths: global agents (~/.codex/agents/) available across all projects and project-specific agents (.codex/agents/) with higher precedence
Isolated context windows per subagent preventing cross-task contamination and maintaining clarity in primary conversations
Explicit delegation workflow patterns for parallel agent execution, sequential coordination, and multi-stage review processes
Domain-specific experts for 30+ programming languages, frameworks (React, Next.js, Django, Spring Boot), and platforms (AWS, Azure, Kubernetes)
Specialized workflow agents including code reviewers, security auditors, performance engineers, and multi-agent orchestrators
Agent installer subagent for browsing and installing additional agents directly from the GitHub repository

Use Cases

  • 01Code review workflows with parallel reviewer, security-auditor, and docs-researcher subagents providing comprehensive analysis
  • 02Bug investigation using code-mapper to trace ownership, browser-debugger to reproduce issues, and specialized language experts to propose fixes
  • 03API development leveraging api-designer for schema design, backend-developer for implementation, and security-auditor for vulnerability scanning
  • 04Infrastructure automation with terraform-engineer, kubernetes-specialist, and cloud-architect subagents coordinating deployment strategies
  • 05Compliance auditing using gdpr-ccpa-compliance, hipaa-compliance, and security-auditor to verify regulatory requirements
  • 06Legacy modernization projects employing legacy-modernizer, refactoring-specialist, and language-specific experts to update codebases

Related Agents

View more

awesome-codex-subagents — FAQ

What are Codex subagents and how do they work?+

Codex subagents are specialized AI assistants defined in .toml files that provide task-specific expertise when explicitly delegated by the user. Each subagent operates in its own isolated context window with custom instructions, preventing the main conversation from becoming cluttered while delivering superior performance on specialized tasks.

How do I install and use Codex subagents from this collection?+

Clone the repository, copy the desired .toml files from the categories folders into either ~/.codex/agents/ (for global availability) or .codex/agents/ (for project-specific use), then restart your Codex session. You must explicitly delegate tasks to subagents in your prompts; Codex does not auto-spawn them.

Which AI clients and editors support these Codex subagents?+

These subagents are designed specifically for Codex, the AI coding assistant. They use Codex-native .toml format and follow the official Codex subagent documentation and directory conventions.

Do I need API keys or special prerequisites to use these subagents?+

You need a working Codex installation with access to the specified models (GPT-5.4, GPT-5.3-codex-spark). No additional API keys are required beyond your Codex configuration. Individual subagents may reference external tools or services in their workflows.

Are these Codex subagents free to use?+

The subagent definitions are open-source under MIT License and free to use. However, running them requires a Codex subscription with access to the underlying AI models, which may have associated costs.

What's the difference between global and project-specific subagent installation?+

Global subagents installed in ~/.codex/agents/ are available across all projects, while project-specific subagents in .codex/agents/ only work in that repository and take precedence over global ones in naming conflicts.

How do I install awesome-codex-subagents?+

Open the source repository on GitHub and follow its README. awesome-codex-subagents is a agent — MCP Agents Market links you directly to the official repo.

Is awesome-codex-subagents free?+

awesome-codex-subagents is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.

Related searches