247-claude-code-remote

The 247-claude-code-remote skill allows users to access Claude Code from any location, providing a secure connection through Tailscale. This skill is particularly useful for developers and AI practitioners who need to provision virtual machines with Fly.io. It seamlessly integrates with popular AI frameworks such as Gemini, Codex, and OpenCode, making it a versatile tool in the AI automation landscape. One of the key benefits of this skill is its ability to facilitate workflow automation by enabling remote access to Claude Code. This means that developers can work on projects from anywhere, reducing the need for physical presence in an office or a specific location. While the exact time savings are not quantified, the flexibility and ease of access can significantly enhance productivity, allowing teams to collaborate more effectively regardless of their geographical locations. This skill is ideal for developers, product managers, and AI practitioners who are looking to streamline their workflows and enhance their project management capabilities. By utilizing the 247-claude-code-remote skill, teams can ensure that they have continuous access to necessary tools and resources, which is crucial in fast-paced development environments. Practical use cases include remote debugging, collaborative coding sessions, and real-time project updates, all of which can be conducted securely and efficiently. With an intermediate implementation difficulty and a setup time of approximately 30 minutes, this skill is accessible for those with a basic understanding of AI automation tools. It fits seamlessly into AI-first workflows by enabling users to leverage AI capabilities without being tethered to a specific location, thus promoting a more agile and responsive development process.

[{"step":"Provision the VM on Fly.io.","action":"Run `flyctl launch` with your desired specifications (e.g., `flyctl launch --name claude-remote-vm --region iad --image flyio/ubuntu:22.04 --vm-size shared-cpu-4x`). Ensure the VM size and region align with your workload needs (e.g., shared-cpu-4x for moderate tasks, dedicated-cpu-8x for heavy workloads).","tip":"Use `flyctl regions list` to check available regions and `flyctl vm sizes` to compare VM options."},{"step":"Install and configure Tailscale.","action":"SSH into the VM (`flyctl ssh console`) and install Tailscale (`sudo apt install -y tailscale`). Authenticate using an auth key (`sudo tailscale up --authkey [YOUR_AUTH_KEY] --hostname [VM_NAME]`). Verify the connection with `tailscale status`.","tip":"Generate an auth key in your Tailscale admin console with ephemeral or reusable settings. Use `--hostname` to make the VM easily identifiable in your Tailscale network."},{"step":"Set up your IDE tool (e.g., VS Code Server).","action":"Install your preferred IDE server (e.g., `curl -fsSL https://code-server.dev/install.sh | sh` for VS Code Server). Start the server with `code-server --auth none --bind-addr 0.0.0.0:8080 &` to allow remote access.","tip":"For security, use `--auth password` or integrate with your identity provider. Restrict access via firewall rules (`ufw allow 8080/tcp`)."},{"step":"Connect to the remote environment.","action":"Use the `247-claude-code-remote` CLI to connect (`claude-code-remote --endpoint https://claude.ai --vm-name [VM_NAME]`). Access the IDE via the Tailscale IP (e.g., `http://[TAILSCALE_IP]:8080`).","tip":"Test the connection from both mobile and desktop devices to ensure cross-device compatibility. Use `claude-code-remote --debug` to log connection issues."},{"step":"Document and automate the setup.","action":"Save the setup steps in a Markdown file (e.g., `claude-remote-setup.md`) and create a `fly.toml` config file for reproducible deployments. Share the environment with your team by inviting their Tailscale emails.","tip":"Use `flyctl deploy` to update the VM configuration from your `fly.toml` file. Schedule daily snapshots (`flyctl snapshot create`) for backup."}]