Agent skill

Handoff

by Szotasz in Szotasz/marveen

Generate a HANDOFF.md context transfer document for session continuity.

MITAuto-check: notesAgent Workflows

Install Handoff

skills CLI
$ npx skills add Szotasz/marveen --skill handoff -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install Szotasz/marveen handoff --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/Szotasz/marveen.git skills-src && mkdir -p .claude/skills && cp -r skills-src/seed-skills/handoff .claude/skills/handoff && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
handoff
GitHub stars
117
Token cost
~1.5k tokens
SKILL.md length
408 words
Files
1
Skills in repo
18
Repo updated
First seen
Licence
MIT

At a glance

Generate a HANDOFF.md context transfer document for session continuity.

  • Works in 4 steps: Gather context → Generate HANDOFF.md → Deliver → …
  • Switching sessions
  • SKILL.md covers When to use, Arguments, Procedure and Pitfalls, plus 1 more section
  • Calls curl and python3

What it does

Handoff is an agent skill from Szotasz/marveen. Generate a HANDOFF.md context transfer document for session continuity. Use when switching sessions, handing off to another agent, or preserving complex task context before a context window reset. Trigger on "/handoff" command or "handoff:" prefix in inter-agent messages.

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Agent Workflows, covering Context engineering. The repository describes itself as: AI csapatod, ami fut amíg te alszol. The licence is MIT.

When your agent uses it

  • Switching sessions
  • Handing off to another agent
  • Preserving complex task context before a context window reset
  • /handoff command

Example prompts

  • “/handoff”
  • “handoff:”
  • “/handoff”

Requirements

  • Python 3

Workflow steps

4 steps, taken from the step headings in SKILL.md.

  1. Gather context
  2. Generate HANDOFF.md
  3. Deliver
  4. Confirm

What it can do on your machine

Read from SKILL.md and the folder at commit e1ac6cc. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • curl
    • python3

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use curl, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Handoff loads about 1.5k tokens when it runs. Until then it costs about 70 tokens; SKILL.md has 408 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~70
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NoteMentions a .env fileSKILL.md:41
    PORT="$(sed -n 's/^WEB_PORT=//p' .env 2>/dev/null | head -1 | tr -d '"')"; PORT="${PORT:-3420}"
  • NoteMentions a .env fileSKILL.md:137
    - Do NOT include secrets, tokens, or .env values in the handoff

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Szotasz/marveen at commit e1ac6cc, republished under its MIT licence (© Szotasz). 408 words, ~1,548 tokens.

Download SKILL.mdSave it as .claude/skills/handoff/SKILL.md (or your agent's skills folder).
name
handoff
description
Generate a HANDOFF.md context transfer document for session continuity. Use when switching sessions, handing off to another agent, or preserving complex task context before a context window reset. Trigger on "/handoff" command or "handoff:" prefix in inter-agent messages.

Handoff -- Session Context Transfer

When to use

  • You are about to hit context limits and need to preserve task state
  • A task needs to continue in a fresh session (yours or another agent's)
  • Inter-agent delegation of a complex, multi-step task
  • User explicitly says /handoff or asks to "save context for later"
  • Before a /checkpoint when the task is too complex for 3-5 bullet points

Arguments

ArgumentRequiredDescription
purposeYESWhat the next session should do with this context
targetnoAgent name (e.g. target=dev2) -- sends via inter-agent message instead of writing file
outputnoFile path override (default: project root HANDOFF.md)

Examples:

  • /handoff purpose="Continue the Pipedrive connector PR review and address CI failures"
  • /handoff purpose="Finish the scheduler forceSend implementation" target=dev2
  • /handoff purpose="Debug the auth redirect loop" output=/tmp/handoff-auth.md

Procedure

1. Gather context

Collect data from these sources (skip any that return empty):

bash
# Active kanban cards (assigned to current agent or recently touched).
# Via the dashboard API, NOT the sqlite3 CLI: sqlite3 (and jq) are absent on a
# stock Linux install -- measured on two live hosts 2026-08-04, where the CLI
# call died with exit 127 while python3 was present on both.
AGENT_ID="$(echo $BOT_NAME | tr '[:upper:]' '[:lower:]')"
PORT="$(sed -n 's/^WEB_PORT=//p' .env 2>/dev/null | head -1 | tr -d '"')"; PORT="${PORT:-3420}"
curl -s -H "Authorization: Bearer $(cat store/.dashboard-token)" \
  "http://localhost:$PORT/api/kanban" | AGENT_ID="$AGENT_ID" python3 -c "
import json,os,sys
me=os.environ.get('AGENT_ID','')
rows=[c for c in json.load(sys.stdin)
      if not c.get('archived_at') and ((c.get('assignee') or '').lower()==me or c.get('status')=='in_progress')]
rank={'urgent':0,'high':1,'normal':2,'low':3}
rows.sort(key=lambda c:(rank.get(c.get('priority'),9), -(c.get('updated_at') or 0)))
for c in rows[:10]:
    print(c['id'], '|', c.get('status'), '|', c.get('priority'), '|', (c.get('assignee') or '-'), '|', c.get('title'))
"

# Hot memories from last 24h
curl -s -H "Authorization: Bearer $(cat store/.dashboard-token)" \
  "http://localhost:$PORT/api/memories?agent=$AGENT_ID&category=hot&limit=10"

# Recent warm memories (project context)
curl -s -H "Authorization: Bearer $(cat store/.dashboard-token)" \
  "http://localhost:$PORT/api/memories?agent=$AGENT_ID&category=warm&limit=5"

# Today's daily log
DATE=$(date +%Y-%m-%d)
curl -s -H "Authorization: Bearer $(cat store/.dashboard-token)" \
  "http://localhost:$PORT/api/daily-log?agent=$AGENT_ID&date=$DATE"

Also include from your current conversation context:

  • The last significant user/peer messages and decisions
  • Any error patterns or debugging findings
  • File paths and line numbers you were working on
  • Git branch, uncommitted changes, open PRs
2. Generate HANDOFF.md

Structure with exactly these 5 sections:

markdown
# Handoff: {purpose}

Generated: {ISO timestamp}
From: {agent name}
To: {target agent or "next session"}

## Goal
{What the overall task is trying to accomplish. 2-3 sentences max.}

## Current Progress
{What has been done so far. Bullet list with specifics:
- File paths changed
- PRs opened (with URLs)
- Kanban card IDs and their status
- Key decisions made}

## What Worked
{Approaches, tools, or patterns that succeeded:
- Specific commands or API calls that gave good results
- Architecture decisions that held up
- Workarounds that solved blockers}

## What Didn't Work
{Dead ends, failed approaches, gotchas:
- Commands or approaches that failed and WHY
- Assumptions that turned out wrong
- Edge cases discovered}

## Next Steps
{Concrete, actionable items for the receiving session:
1. First thing to do (most specific)
2. Second thing
3. ...
Keep each step concrete enough to execute without asking questions.}
3. Deliver

File mode (default): Write HANDOFF.md to the project root (or output path).

Inter-agent mode (target= specified): Send the full HANDOFF.md content as an inter-agent message:

bash
curl -s -X POST http://localhost:$PORT/api/messages \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $(cat store/.dashboard-token)" \
  -d "{\"from\":\"$AGENT_ID\",\"to\":\"TARGET\",\"content\":\"[HANDOFF] purpose: ... \n\n$(cat HANDOFF.md)\"}"
4. Confirm

Report to the user/caller:

  • Where the handoff was written (file path or inter-agent message ID)
  • Summary: how many kanban cards, memories, and log entries were included
  • The purpose line for quick reference
Show full SKILL.md (161 more words)Show less

Pitfalls

  • Do NOT include secrets, tokens, or .env values in the handoff
  • Do NOT include full file contents -- use paths and line numbers
  • Keep it under 3000 words -- the receiving session needs room to work
  • If target= agent is not running (check tmux), warn and fall back to file mode
  • The handoff is a snapshot -- it goes stale. Include the timestamp prominently

Relation to other persistence mechanisms

MechanismScopeHandoff uses it as
checkpointSession summary (SQLite)Source: pulls recent checkpoint data
DREAM.mdNightly consolidationNot directly -- too high-level
hot memoryActive task stateSource: includes active hot memories
warm memoryStable project contextSource: includes relevant warm context
kanbanTask trackingSource: includes assigned/active cards
daily logChronological recordSource: includes today's log entries

The handoff READS from these systems but does not REPLACE them. After a handoff, the receiving session should still check the live state of kanban/memory -- the handoff is a starting-context accelerator, not the source of truth.

© Szotasz, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in seed-skills/handoff of Szotasz/marveen.

Open the folder on GitHubat commit e1ac6cc

Compare with similar skills

Handoff next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Handoff compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Handoff this skillSzotasz/marveen117—~1.5kAutomated safety check: NotesMIT
Context Mode Output Sandboxmksglu/context-mode26k—~4.1kAutomated safety check: PassCustom licence
Memori Long-Term MemoryMemoriLabs/Memori17k—~2kAutomated safety check: NotesCustom licence
Picoclaw Skill Creatorsipeed/picoclaw30k—~4.4kAutomated safety check: PassMIT
ccc Semantic Code Searchcocoindex-io/cocoindex-code2.8k—~938Automated safety check: PassApache-2.0
Context Mode for Antigravity CLImksglu/context-mode26k—~850Automated safety check: PassCustom licence

Similar skills

  • Context Mode Output Sandbox

    mksglu/context-mode

    Routes large command, file, API and browser output through context-mode tools so only the needed result enters the agent's context, instead of dumping it via Bash.

    26k GitHub stars~4.1k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Memori Long-Term Memory

    MemoriLabs/Memori

    Connects Claude Code to Memori Cloud for long-term memory, recalling stored context before substantive replies and saving new context afterward.

    17k GitHub stars~2k tokensUpdated 7 days ago
    Agent WorkflowsAuto-check: notes
  • Picoclaw Skill Creator

    sipeed/picoclaw

    Guidance for creating, updating and reviewing Picoclaw skills, from the SKILL.md structure to organizing bundled scripts, references and assets.

    30k GitHub stars~4.4k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • ccc Semantic Code Search

    cocoindex-io/cocoindex-code

    Semantic code search and index management with the ccc CLI: the agent initializes, indexes and queries the project by concept, filtering by language or path.

    2.8k GitHub stars~938 tokensUpdated 3 days ago
    Agent WorkflowsAuto-check passed
  • Routing rules for using context-mode MCP tools in Antigravity CLI: sandboxed code runs, file analysis, indexed search and web fetches that keep large output out of the conversation.

    26k GitHub stars~850 tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Token Optimizer

    alexgreensh/token-optimizer

    Audit a Claude Code or Codex setup for context-window waste, then fix it and measure the savings.

    2.5k GitHub stars~3.6k tokensUpdated yesterday
    Agent WorkflowsAuto-check passed

More from Szotasz/marveen

All 18 skills in this repo
  • Fleet Helper

    Szotasz/marveen

    Shared, dependency-free Python helpers for the agent fleet - dashboard API (memory, messages, kanban), Telegram MarkdownV2 escaping, and rule-based Mail.app triage.

    117 GitHub stars~1.7k tokensUpdated today
    Auto-check passed
  • GitHub PR Rebase Merge

    Szotasz/marveen

    Merge a stack of GitHub PRs sequentially when they share files and will cause cascading conflicts.

    117 GitHub stars~1.8k tokensUpdated today
    Auto-check passed
  • Retrospective

    Szotasz/marveen

    Analyze the current session for improvement opportunities in skills, memory, and workflow.

    117 GitHub stars~1.5k tokensUpdated today
    Auto-check passed
  • A Marveen Bridge és a "Szolgáltatás-lapok" ajánlása és elmagyarázása a gazdának.

    117 GitHub stars~1.4k tokensUpdated today
    Auto-check: warnings
  • Skill Factory

    Szotasz/marveen

    Turn any workflow, conversation, or demonstrated process into a reusable SKILL.md.

    117 GitHub stars~1.3k tokensUpdated today
    Auto-check passed
  • Skill Management

    Szotasz/marveen

    List, inspect, patch, or delete skills from ~/.claude/skills/.

    117 GitHub stars~1.9k tokensUpdated today
    Auto-check passed

Categories

Questions about Handoff

What does Handoff do?

Generate a HANDOFF.md context transfer document for session continuity. Handoff is an agent skill from Szotasz/marveen.md context transfer document for session continuity.

When should I use Handoff?

Handoff fits situations like: switching sessions; handing off to another agent; preserving complex task context before a context window reset; /handoff command.

How do I install Handoff in Claude Code?

Run `npx skills add Szotasz/marveen --skill handoff -a claude-code`. Or copy the skill folder (seed-skills/handoff in Szotasz/marveen) into .claude/skills/handoff in your project. Claude Code loads it when a task matches its description.

How do I install Handoff in Codex?

Run `npx skills add Szotasz/marveen --skill handoff -a codex`. Or copy the skill folder (seed-skills/handoff in Szotasz/marveen) into .agents/skills/handoff in your project. Codex loads it when a task matches its description.

Can I use Handoff in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add Szotasz/marveen --skill handoff -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/handoff, .gemini/skills/handoff, .github/skills/handoff and .opencode/skills/handoff in your project.

What does Handoff need to run?

Going by SKILL.md and its folder, Handoff needs the command-line tools its instructions call (curl and python3). Our summary lists: Python 3.

Does Handoff access the network?

SKILL.md contains no URLs. Its commands use curl, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Handoff safe to install?

Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.

What licence does Handoff use?

Handoff is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Handoff use?

About 1.5k tokens (SKILL.md is roughly 6.2k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Handoff?

Skills that share tags, products or a category with Handoff: Context Mode Output Sandbox (mksglu/context-mode, 26k stars), Memori Long-Term Memory (MemoriLabs/Memori, 17k stars), Picoclaw Skill Creator (sipeed/picoclaw, 30k stars) and ccc Semantic Code Search (cocoindex-io/cocoindex-code, 2.8k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Handoff?

Szotasz (a GitHub user) maintains it in Szotasz/marveen, which has 117 GitHub stars. The repository holds 18 skills in this directory. The repository was last updated on October 10, 2026.

Source: Szotasz/marveen on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.