Install
$ agentstack add skill-seilk-claude-teleport-teleport-init ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
Teleport Init
Set up Teleport for the first time on this machine.
Steps
- Check prerequisites: Run
node "${CLAUDE_PLUGIN_ROOT}/dist/cli.js" context. Parse the JSON result.
- If
auth.ghInstalledis false: show install command based onauth.os(darwin:brew install gh, linux:sudo apt install gh). STOP. - If
auth.authenticatedis false: showgh auth logininstructions. STOP. - Store
auth.usernameandmachine.alias.
- Confirm machine name: Show the detected
machine.aliasto the user. Ask if they want to keep it or rename. If an argument was provided, use that instead.
- Create hub repo: Run
node "${CLAUDE_PLUGIN_ROOT}/dist/cli.js" hub-init --clone-to /tmp/claude-teleport-hub. Parse result.
- If
createdis false: inform user "Hub already exists at . Using existing." - If
createdis true: inform user "Created private hub at ." - Store
usernameandlocalPathfrom the result for later use.
- Scan local environment: Run
node "${CLAUDE_PLUGIN_ROOT}/dist/cli.js" scan --claude-dir ~/.claude --output /tmp/teleport-scan.json. Parse the JSON output — it contains asummaryobject with counts per category.
- Present summary: Using the
summaryfrom the scan output, present:
``` Found on this machine:
- Plugins: {summary.plugins}
- Agents: {summary.agents}
- Rules: {summary.rules}
- Skills: {summary.skills}
- Hooks: {summary.hooks}
- Settings keys: {summary.settings}
- MCP configs: {summary.mcp}
```
- Category selection: Use
AskUserQuestionwithmultiSelect: trueto present categories (Plugins, Agents, Rules, Skills, Hooks, Settings — only show categories with items). Then for each selected category, use anotherAskUserQuestionwithmultiSelect: truelisting the individual items, asking "Any items to exclude?"
Important: In the snapshot JSON, most categories (agents, rules, skills, commands, mcp) are arrays of FileEntry objects with a relativePath field. However, settings is a plain object (key-value pairs, not an array) — use Object.keys() to list its items. plugins and marketplaces are arrays of objects with a name field. hooks is an array of objects with name and event fields.
- Secret scan: Run
node "${CLAUDE_PLUGIN_ROOT}/dist/cli.js" secret-scan --snapshot-file /tmp/teleport-scan.json --output /tmp/teleport-secrets.json. Read findings. If any: show each finding with file and line. Confirm exclusion.
- First-push review gate: "This is the first push. Please review the files that will be committed:" Show the file list. Wait for user confirmation.
- Push to hub: Write selections to
/tmp/teleport-selections.json. Runnode "${CLAUDE_PLUGIN_ROOT}/dist/cli.js" hub-push --hub-path --machine --username --snapshot-file /tmp/teleport-scan.json --selections-file /tmp/teleport-selections.json. This writes configs undermachines//in the hub, generatesregistry.yamlon main, and creates an agent-friendlyREADME.md.
- Success: Show "Your Claude Code setup has been teleported to . Run
/teleport-pullon other machines to apply."
- Cleanup: Remove temp files.
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: seilk
- Source: seilk/claude-teleport
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.