Turns a codebase, its docs, schemas and PDFs into a queryable knowledge graph for AI coding assistants, built locally with tree-sitter AST parsing.
markfulton/ai-employees
Open source AI Employees: eight scheduled business-role kits with 60 routines that run on your own machine through an AI agent like Claude Code.
- Stars
- 635
- Forks
- 171
- Open issues
- 8
- Last commit
- 3 days ago
What is it?
AI Employees is a set of plain-text folders, each covering one business role such as GTM Engineer, SEO/AEO, Social Media, Ad Manager, Sales, Customer Satisfaction, or Chief of Staff. Each folder holds a role definition, a schedule, and one subfolder per routine, and it runs on the user's own machine through an AI agent such as Claude Code, with support for about a dozen other agent harnesses.
The routines are designed to drive a browser and a PC as the user would, signed in with the user's own accounts. Most actions such as sending, posting, submitting, publishing or spending are drafted or staged and held for the user's click, and the user releases channels explicitly in a RELEASES.md file. Each morning the employee produces a brief listing finished work, items waiting on the user, and blocked tasks.
The project is MIT-licensed, supports Windows, macOS and Linux, and can be installed by download, through an npm package, or as a Claude Code plugin. Upgrades are designed to leave locally edited files untouched.
Key features
- 01Eight business-role kits with 60 scheduled routines in total
- 02Runs on Claude Code and about a dozen other AI agent harnesses
- 03Drives the browser and PC using the user's own signed-in accounts
- 04Drafts and stages sends, posts and spend for the user to approve
- 05Produces a daily morning brief of finished, waiting and blocked work
- 06Upgrades keep locally edited files intact
Use cases
- Running marketing, outbound sales and SEO routines for a small business
- Automating daily inbox triage and customer follow-up drafts
- Monitoring ad account performance and preparing weekly change lists
- Giving a solo founder a recurring operations checklist with human approval gates
Quick start
- 1
Install the GTM Engineer kit from the command line
npx ai-employees hire gtm-engineer --to <folder> - 2
Open your AI agent in that folder and say "Install the GTM Engineer from this folder."
Things to know
- You need an AI agent you are logged in to, and a browser signed in to the accounts the employees should read.
- The kit folder must sit outside cloud sync folders such as OneDrive, Dropbox, Google Drive or iCloud.
Topics
FAQ
- What is ai-employees?
- Open source AI Employees: eight scheduled business-role kits with 60 routines that run on your own machine through an AI agent like Claude Code. AI Employees is a set of plain-text folders, each covering one business role such as GTM Engineer, SEO/AEO, Social Media, Ad Manager, Sales, Customer Satisfaction, or Chief of Staff. Each folder holds a role definition, a schedule, and one subfolder per routine, and it runs on the user's own machine through an AI agent such as Claude Code, with support for about a dozen other agent harnesses.
- Is ai-employees free to use? What license does it have?
- ai-employees is open source under the MIT license. Check the license terms before using it in commercial products.
- Is ai-employees still maintained?
- Actively maintained. The last commit to ai-employees was on Oct 17, 2026.
- What can I use ai-employees for?
- Running marketing, outbound sales and SEO routines for a small business Automating daily inbox triage and customer follow-up drafts Monitoring ad account performance and preparing weekly change lists Giving a solo founder a recurring operations checklist with human approval gates
- How do I install ai-employees?
- Install the GTM Engineer kit from the command line: npx ai-employees hire gtm-engineer --to <folder> Open your AI agent in that folder and say "Install the GTM Engineer from this folder."
- Where can I get ai-employees?
- The source code and installation instructions are on GitHub: https://github.com/markfulton/ai-employees
Summary generated by AI from the project's README. GitHub stats refresh daily.
Related projects
Multi-agent orchestration layer for Claude Code, adding team pipelines, autonomous execution and tmux-based CLI workers.
A browser built for AI agents and you to share: agents run tasks in isolated Spaces while your own tabs and logins stay untouched.
A collection of 74 game-development skill files for AI coding agents, with a router that loads the right engine and task skills.
A set of installable skills that give AI coding agents guidance on motion design, After Effects, Remotion video, logo animation, and motion backgrounds.
A set of 23 slash-command skills and power tools that turn Claude Code into a virtual team of CEO, designer, eng manager, reviewer, QA and release roles.