skills.rest
letta-ai
letta-aiOFFICIAL
Explore all skills by this author →
Website𝕏@letta_aiGitHub

teleporting-between-environments

Moves agent conversations between Cloud, Desktop, and remote environments while preserving memory.

3.2k|385|Updated Oct 25, 2025
Search Tags:#agent handoff#remote execution#file transfer#cloud sandbox#teleport#letta cli#environment migration
One-click install
npx skills add https://github.com/letta-ai/letta-code --skill teleporting-between-environments
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: teleporting-between-environments
Source: https://github.com/letta-ai/letta-code/tree/main/src/skills/builtin/teleporting-between-environments
Command: npx skills add https://github.com/letta-ai/letta-code --skill teleporting-between-environments

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Continuing an agent conversation on a different computer normally loses context or requires manual re-setup. This Skill coordinates moving the current conversation between Cloud sandboxes, Desktop Local, and other Cloud-registered environments while handling the machine-local files and setup that do not transfer automatically.

Core Features & Use Cases

  • Environment Discovery and Targeting: List accessible online Cloud-registered targets with letta teleport list and move the conversation with letta teleport cloud, letta teleport local, or a named environment.
  • Artifact Transfer: Upload and download files between machines using letta sandbox upload and letta sandbox download before the handoff, retaining returned sandbox paths in context.
  • Headless Environment Routing: Run a separate headless turn on Cloud or another environment with letta -p --conversation <id> --environment <target> without moving the current conversation.
  • Use Case: You started a debugging task on your laptop but need more compute. Upload the local artifacts, run letta teleport cloud as the final action, and the conversation resumes in the Cloud sandbox with full memory intact.

Quick Start

Ask the agent to continue this task on Cloud, and it will upload any needed local files and run the teleport command as its final action.

Frequently Asked Questions about teleporting-between-environments

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I move a Letta agent conversation to the Cloud?▼

Upload any local files the task needs with letta sandbox upload, retain the returned sandbox paths, then run letta teleport cloud as the final command. The conversation resumes in the managed Cloud sandbox with memory intact.

How do I return a conversation from Cloud to my local Desktop?▼

Run letta teleport local as the final action to target the online Letta Desktop environment. Desktop Remote Access must be enabled, and if multiple Desktop environments are online, use letta teleport list to pick one explicitly.

Do files and working directories transfer when teleporting between environments?▼

No, filesystem paths, working directories, credentials, and running processes stay on the source machine. You must upload needed artifacts beforehand and re-establish the repository, dependencies, and services on the destination after arrival.

Can I run a one-off headless command on another environment without teleporting?▼

Yes, use letta -p --conversation <id> --environment cloud "<prompt>" or specify an environment name, device ID, or connection ID. This routes only that headless message and does not move the current conversation.

What happens if the teleport target is offline or unsupported?▼

The conversation remains on the source machine and the command reports an offline, stale, unsupported, same-source, or startup error. Surface the error, correct the target availability, and retry only once the destination is online.

Related Skills

Explore More Software Engineering Skills→
VrooliVrooli
54

Landing Page Deploy Setup

Validate LPBS readiness gates before desktop artifact deployment handoff.

Official
Advanced
lihaoze123lihaoze123
2

trellis-start

Initialize coding sessions by loading project context and routing workflows.

Community
Intermediate
MasterleeausMasterleeaus

zylos-memory

Store and organize session context across AI conversations using structured markdown.

Community
Advanced
CaptainCrouton89CaptainCrouton89
25

a name

Run subagents in forked contexts with isolated prompts.

Community
Basic