How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-teleport:teleport-from <github-username><github-username>This skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Import configurations from another user's public teleport repo.
Import configurations from another user's public teleport repo.
Clone public repo: Clone <username>/claude-teleport-public to a temp directory. The repo contains an agent-friendly README.md that describes the repository structure and import instructions.
List machines: Read registry.yaml from the cloned repo to list available machines and their config counts. Use AskUserQuestion (single-select) to let the user pick a machine. Options: each machine name with config counts as description (e.g. "macbook-pro (3 agents, 5 rules)"), plus "main (merged union of all machines)".
Read machine configs: Read files from machines/<selected-alias>/ in the repo. Present categories: "Available: X agents, Y rules, Z skills." File paths inside machines/<alias>/ map directly to ~/.claude/ paths.
User selects: Use AskUserQuestion with multiSelect: true to present available categories. Then for each selected category, use another AskUserQuestion with multiSelect: true listing specific items.
Mandatory content review: For ALL selected files (not just hooks):
rce-scan --file <path> on each.Backup: Run backup --claude-dir ~/.claude. Show backup path.
Apply: Apply selected items using the apply command.
Show result: Display what was applied, any errors, and plugin install instructions.
Cleanup: Remove temp directory.
npx claudepluginhub seilk/claude-teleport --plugin claude-teleportCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.