Agency Agents for Openclaw

A massive workforce of 61 specialized AI agents organized into departments to handle professional engineering, design, and marketing projects.

jerry-guo-mys
v1.0.0
Mar 7, 2026
1
1.9k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install agency-agents-openclaw

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install agency-agents-openclaw using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Agency Agents?

Agency Agents is a high-performance collection of Openclaw Skills designed to simulate a full-scale professional agency. It features 61 distinct AI personas partitioned into eight core departments, including Engineering, Design, Marketing, Product, and Quality Assurance. This structure allows developers and project managers to deploy individual specialists for targeted tasks or utilize the Orchestrator to coordinate complex, multi-stage workflows.

Built to enhance the capabilities of AI coding environments, these Openclaw Skills provide deep expertise in specific technical niches like React development, UX research, and growth hacking. Whether you need a single component built or an entire digital product launched, Agency Agents provides the structured intelligence required to execute at an enterprise level.

Agency Agents Use Cases

  • Launching full-stack web applications by coordinating frontend, backend, and DevOps agents.
  • Designing comprehensive brand systems and UI components using specialized design specialists.
  • Executing multi-platform marketing strategies across TikTok, Twitter, and Reddit.
  • Automating complex QA cycles and performance benchmarking for enterprise software.

How Agency Agents Works

  1. Users trigger the skill by selecting a specific agent, a whole department, or the master orchestrator.
  2. The system loads the designated agent's persona, specialized knowledge base, and quality assurance parameters.
  3. For complex requests, the Orchestrator performs task decomposition, breaking the project into manageable sub-tasks for different specialists.
  4. Agents execute their specific roles (e.g., the backend-architect designs the API while the frontend-developer builds the UI).
  5. Results are passed through a quality check phase where reality-checker agents verify the deliverables against the initial requirements.

Agency Agents Setup

To begin using these Openclaw Skills, you can configure your environment to match your preferred workflow and quality standards.

# Set your default department for quick access
export AGENCY_AGENTS_DEFAULT_DEPARTMENT=engineering

# Configure the QA strictness level (1-5)
export AGENCY_AGENTS_QA_LEVEL=3

# Example: Running a single agent for a specific task
/openclaw skill use agency-agents --agent frontend-developer "Create a responsive login page using Tailwind CSS"

Agency Agents Data Schema & Taxonomy

The skill produces highly structured output to ensure consistency across the agent workforce:

Section Description Content Type
Task Understanding Analysis of the project scope and constraints Markdown Text
Execution Process Step-by-step log of the agent's internal workflow Numbered List
Delivery Assets The actual code, documentation, or strategy files Code Blocks/Files
Quality Check Self-verification results and compliance reports Checklist
Recommendations Suggestions for further optimization and next steps Bullet Points

Agency Agents Advanced Features

  • Multi-agent orchestration for end-to-end project management and task delegation.
  • Departmental collaboration mode to activate a specific group of specialists simultaneously.
  • Dynamic QA levels to control the depth of automated testing and reality checks.
  • Cross-functional feedback loops where testing agents can automatically trigger retries for engineering agents.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*