Sync AI agent skill repositories and install them into a target project.
See which skills are already installed versus available to add.
Package skills for GitHub Copilot, OpenCode, or Antigravity IDE.
Provision skills into a project from the command line instead of the dashboard.
| chrys/agentcraft | bitman-sachs/minimax-m3-tp3-rtx6000 | brandur/heroku-buildpack-mono-build | |
|---|---|---|---|
| Stars | 2 | 2 | 2 |
| Language | Shell | Shell | Shell |
| Last pushed | — | — | 2012-01-27 |
| Maintenance | — | — | Dormant |
| Setup difficulty | easy | hard | moderate |
| Complexity | 2/5 | 5/5 | 2/5 |
| Audience | developer | ops devops | ops devops |
Figures from each repo's GitHub metadata at analysis time.
AgentCraft is a dashboard and command line tool for managing AI agent skills across your software projects. Many coding tools now support skills, reusable instructions that teach an AI assistant how to handle a specific kind of task, and AgentCraft helps you pull those skill definitions from GitHub repositories and install them into whichever project and tool you are working with. It runs as a local web dashboard. From the Configuration tab you register the repositories you want to pull skills from, giving each one a name, its clone URL, a branch, and the path to its skills inside that repo, then click a button to sync and clone or update them. From the Skills tab you point it at a target project folder on your computer and choose which tool style you are packaging for: GitHub Copilot, OpenCode, or an editor called Antigravity IDE. The dashboard then scans your project and splits available skills into two lists, ones already installed and ones you could add. Depending on which tool you choose, AgentCraft copies skill files to the right folder automatically, such as a hidden skills folder for GitHub Copilot or Antigravity IDE, or a plain skills folder for OpenCode. For Antigravity IDE specifically, it can also generate matching custom workflow files. You can select individual skills to apply or remove, and cloned source repositories are kept in a dedicated folder that stays out of your project's own version control. If you prefer not to use the web interface, two shell scripts let you do the same work from the command line: one syncs your configured repositories and updates a table in this project's own README, and the other copies chosen skills into a target project by index number or by name. Getting started only requires installing Node dependencies and starting the development server.
A dashboard and CLI tool for syncing AI agent skill repositories and installing them into your projects.
Mainly Shell. The stack also includes Node.js, Shell, HTMX.
Setup difficulty is rated easy, with roughly 30min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
double-check against the repo, no cap.