handoff

A command that sends the current coding task to Claude Code running in a Sprites.dev cloud sandbox, a remote computer environment for running code.

In plain words
What is it for?
Use it to hand off a task, choose a Claude model and spending limit, and receive information for resuming the remote session.
Why use it?
It lets long-running or background work continue remotely while keeping the local project separate through a Git worktree, an isolated copy of the files.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/superfly/claude-plugins/handoff
Clone the repo
git clone --depth 1 https://github.com/superfly/claude-plugins
Per session 15 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,027 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00015 $0.01027
Opus 5 $0.00008 $0.00513
Sonnet 5 $0.00003 $0.00205
Haiku 4.5 $0.00002 $0.00103

Measured 2d ago against content hash 317a0b1689bf, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

handoff scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

sprite-handoff/commands/handoff.md · 139 lines

How it starts

The opening of the file, as written. The whole thing — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Hand Off to Remote Claude Code

Transfer the current work to a Claude Code instance running on a sprite in the cloud. This enables background processing, parallel work, or offloading long-running tasks.

Workflow

  1. Ensure sprite is ready

    • Check for active sprite with sprite use
    • If none, create one or ask user to run /sprite:create
  2. Generate handoff context

    • Create a summary of the current conversation/task
    • Combine with user's task description
    • Format as a prompt for the remote Claude
  3. Run the handoff using the CLI tool

    $CLAUDE_PLUGIN_ROOT/bin/sprite-handoff run <sprite-name> "<prompt>" [flags]
    

    The run command handles everything:

    • Creates git worktree for session isolation
    • Syncs all git-tracked files
    • Creates initial commit
    • Retrieves Claude OAuth token
    • Launches Claude Code on the sprite

    Available flags:

    • --model <model> - Specify model (sonnet, opus, haiku)
    • --max-budget <amount> - Set spending limit in USD
    • --source <dir> - Source directory (default: current dir)
    • -v - Verbose output
  4. Return session info

    • The command outputs the session ID
    • Explain how to resume: sprite-handoff run <sprite> "<prompt>" --resume <session-id>
    • Explain how to pull changes: sprite-handoff pull <session-id>

Handoff Prompt Format

The prompt passed to the run command should include context:

## Context
[Summary of conversation and what was being worked on]

## Project State
[Key files, recent changes, current status]

## Task
[User's task description - what to continue working on]

## IMPORTANT: Commit Your Changes
You MUST commit all changes before finishing. The user will pull your work
using `git diff`, so uncommitted changes will be lost.

When done, run:
  git add -A && git commit -m "Description of changes"

## Instructions
- Work autonomously on this task
- Commit all changes with a descriptive message when complete
- Write a summary of what was done to HANDOFF_RESULT.md

Read the full file on GitHub · 139 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 139 lines · 15 tokens per session scan A 317a0b1689bf

Subscribe to this mod's changes

handoff is a command published in the GitHub repository superfly/claude-plugins (2 stars, last pushed 7mo ago), licensed MIT. It adds 15 tokens to every session and 1,027 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.