claude-code:handoff

claude-code:handoff is a skill for Claude Code, Codex from bendrucker/claude. It costs 89 tokens per session (1,287 once invoked), scanned A, original, MIT.

Instructions for handing the current coding task to a new background agent. The new agent receives a written summary of the goal, current state, decisions, next steps, files, and constraints.

In plain words
What is it for?
Use it when a task is well defined and should continue in the background, in another session, or with a fresh agent.
Why use it?
It lets work continue while you step away or track it separately, without requiring the new agent to know the earlier conversation.

Skill for Claude CodeCodex

Part of the claude-code plugin — 15 skills shipped together

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 skills/bendrucker/claude/handoff
Any agent
npx skills add bendrucker/claude --skill handoff
Clone the repo
git clone --depth 1 https://github.com/bendrucker/claude

Made for: Claude Code, Codex.

Or install claude-code, the plugin that ships this one along with the rest of its 15 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for claude-code:handoff

README.md
[![agentmods](https://agentmods.dev/badge/skills/bendrucker/claude/handoff.svg)](https://agentmods.dev/skills/bendrucker/claude/handoff)
Your own site
<a href="https://agentmods.dev/skills/bendrucker/claude/handoff"><img src="https://agentmods.dev/badge/skills/bendrucker/claude/handoff.svg" alt="Measured on agentmods" height="20"></a>
Per session 89 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,287 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 $0.00089 $0.01287
Opus 5 $0.00044 $0.00643
Sonnet 5 $0.00018 $0.00257
Haiku 4.5 $0.00009 $0.00129

Measured yesterday against content hash b157e09a2013, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

claude-code: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 yesterday.

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.

plugins/claude-code/skills/handoff/SKILL.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Handoff

Hand the current work to a fresh background agent seeded with a summary of this conversation. The agent starts immediately in the current directory and returns control, so you can walk away and check on it later. Nothing is written to disk: the summary becomes the new agent's opening prompt. When the work is well-scoped and the user wants to step away or parallelize, offer the handoff.

Write the Summary

Synthesize the conversation into a prompt that lets an agent with zero prior context resume without asking questions. Cover:

  • Goal: what the work is trying to achieve, in one or two sentences.
  • State: what is done, what is in progress, what is untouched.
  • Decisions: choices already made and the reasoning, so the agent does not relitigate them.
  • Next steps: the concrete actions to take, in order.
  • Anchors: the branch, PR, issue, plan doc, or key files by path or URL. Point at these rather than restating them.
  • Suggested skills: a short list of skills the agent should invoke for this work (e.g. pull-request:create, review:code, a project skill). This is the biggest lever on whether the agent picks up your conventions.
  • Constraints: anything that would surprise a fresh agent, such as test commands, gotchas, or things not to touch.

Keep it dense and specific. If the user passed an argument, treat it as the focus for the next session and shape the summary around it. With no argument, summarize the conversation as a whole.

Redact

The summary becomes the agent's prompt, so strip anything sensitive before launching: API keys, tokens, passwords, connection strings, PII. Reference where a secret lives (an env var, a secrets manager) rather than its value.

Size the Run

The agent runs unattended, so choose its horsepower and autonomy up front. Propose all three to the user with the launch command and your reasoning, and let them override.

Model and Effort

Match --model and --effort to the task. A well-specified, mechanical task runs fine on --model sonnet at --effort medium. Reserve --model opus with --effort high (or xhigh/max) for open-ended, high-stakes, or reasoning-heavy work. --effort takes low, medium, high, xhigh, or max. --model takes an alias (opus, sonnet, haiku, fable) or a full model name.

Read the full file on GitHub · 76 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. yesterday Changed · +1 lines b157e09a2013
  2. 5d ago First seen · 75 lines · 89 tokens per session scan A da4604187ccb

Subscribe to this mod's changes

claude-code:handoff is a skill published in the GitHub repository bendrucker/claude (16 stars, last pushed yesterday), licensed MIT. It adds 89 tokens to every session and 1,287 once invoked, about $0.0004 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 2026-08-30.

Related

Other skills, from other repositories

create-site

Creates a new Power Pages code site (SPA) using React, Angular, Vue, or Astro. Guides through the full process from initial concept to deployed site: requirements discovery, scaffolding, component planning, design, implementation, validation, and deployment. Use when the user wants to create, build, or scaffold a new…

microsoft/power-platform-skills · 73 tokens

google-ads-audit

Google Ads account audit and business context setup. Run this first — it gathers business information, analyzes account health, and saves context that all other ads skills reuse. Trigger on "audit my ads", "ads audit", "set up my ads", "onboard", "account overview", "how's my account", "ads health check", "what should…

nowork-studio/notfair-plugin · 114 tokens

review

5-pass structured code review — correctness, security, performance, readability, consistency.

SethGammon/Citadel · 17 tokens

alive:system-upgrade

Upgrade ALIVE to the current version. Handles v1/v2/v3.x source states, multi-surface aware (alive-mcp / Hermes / Codex), retroactive version detection, partial-failure resume, dry-run previews, and rollback inspection.

alivecontext/alive · 57 tokens

codex-test-gen

Generate unit tests for specified functions using Codex MCP.

sd0xdev/sd0x-harness · 15 tokens

pm-verify-release-v1

Post-release acceptance verification, grade a shipped release against the spec that promised it. Use when someone says "verify the release", "did the release meet its acceptance scenarios", "post-release verification", "execute the measurement plan", "week-one metrics check", or hands a shipped scope that needs…

naderelewa/Product-to-Prod · 189 tokens