@aishelf/service

@aishelf/service

A lightweight background service that powers AIShelf clients - including the VS Code extension and other future clients. It acts as a centralized source of workflows, skills, prompts, and rules for all your AI tools across your local machine, handling authentication and registry management so that AIShelf tools can securely access your shared AI resources.

For full installation instructions and documentation, visit aishelf.dev/docs/installation.

Installation

Docker (Recommended)

Docker is the recommended way to run the AIShelf service. It handles auto-updates and restarts automatically when Docker starts.

docker rm -f aishelf-service
docker run -d \
  --name aishelf-service \
  --restart=unless-stopped \
  -p 5314:5314 \
  -v ~/.aishelf:/root/.aishelf \
  aishelf/service:latest

The first command removes any existing container (safe to run even if none exists). The second starts a fresh one.

See aishelf.dev/docs/installation#docker for more details.

npm global install (Beta)

Note: This installation method is currently in beta. Docker is recommended if available.

npm install -g @aishelf/service
aishelf service install

See aishelf.dev/docs/installation#npm for more details.

Usage

aishelf service start    # start the service
aishelf service stop     # stop the service
aishelf service restart  # restart the service
aishelf --version        # check version
aishelf --help           # get help

How it works

Once running, @aishelf/service exposes a local HTTP API on localhost:5314. AIShelf clients (IDE extensions, browser extensions, etc.) communicate with it to:

• Authenticate with your GitHub account
• Connect and sync AIShelf registries
• Read and manage your shared AI resources locally

You typically don't need to interact with this service directly — it's managed automatically by the AIShelf clients you use.

API

API documentation coming soon. Visit aishelf.dev for updates.

License

MIT