A library of 266 preconfigured AI expert roles spanning 20 departments, designed for use across 18 AI programming tools. It provides agent definitions with distinct identities, workflows, and deliverables, along with scripts to convert these definitions into tool-specific formats.
Project overview
It provides structured expert agent definitions that specify thinking processes, workflows, and deliverables rather than simple prompts, and includes a conversion script to adapt these definitions for various AI tools.
Project type
AI Agent · Prompt Engineering
Use cases
Documents & Office · Coding & Development
Deployment
Refer to project documentation
License
MIT
Best for
Developers and creators looking for structured instruction sets covering ai-coding, content-writing, and office-automation tasks.
Key capabilities
A collection of 266 AI expert roles covering 20 departments, each with a distinct identity, workflow, and deliverables.
Includes a script to convert agent definitions into the specific formats required by various AI tools.
Limitations and risks
GPU requirements, minimum hardware, OS compatibility, model requirements, and runtime databases are not documented.
Telemetry, data boundary, and cost dependencies are not documented.
Getting started
Setup difficulty is unclear. The project provides one-click install scripts and supports copying files manually.
Run ./scripts/install.sh to install an agent to an AI tool. Coding requirements are optional.
Alternatives and comparisons
Provides structured instruction sets for tasks like goal definition, disk analysis, and documentation alignment for 40+ AI agents.
A curated directory of premium Agent Skill resources categorized by use cases.
A practical tutorial on prompt engineering concepts across text, code, and image generation domains.