Handoff logo

Handoff

Community
Mathews-Tom
handoff

Produces and refreshes `.docs/handoff.md`, a 200-line session-continuity runbook for coding agents. Captures git state, working-tree changes, last validation commands, session decisions, blockers, resume checklist, references, and greenfield scaffolds. Triggers on: "update handoff", "session handoff", "refresh handoff", "/handoff", "/handoff init", "resume checklist", "cold-start primer". Use this skill when ending, pausing, resuming, or transferring in-flight project work across agent sessions.

Overview

PublisherMathews-Tom
Repositoryarmory
Skill namehandoff
Stars
318
Forks
47
Bundled files
4
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 4 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Mathews-Tom on GitHub. Read the source before you install it.

Installation

Install the Handoff AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Mathews-Tom/armory.git /tmp/armory
mkdir -p .claude/skills
cp -r /tmp/armory/skills/handoff .claude/skills/handoff
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Handoff in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Handoff on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Handoff is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Handoff

Maintains .docs/handoff.md as the project-local cold-start primer for the next coding agent or human developer. The file captures transient session state only: what changed, what was decided, what is blocked, and exactly how to resume.

Reference Files

FileContentsLoad When
references/schema.mdVerbatim handoff schema and line-budget rulesAlways before writing
references/authority-boundaries.mdBoundary between handoff, memory, CLAUDE.md, plans, and gitAlways before writing

When To Use

Use handoffUse another surface
In-flight work, uncommitted edits, current branch, blockers, resume stepsgit log for committed history
Session-scoped decisions and failed approachesMemory or ADRs for durable project facts
Immediate validation state and exact next commandCLAUDE.md for stable setup instructions
Current roadmap cursorPlan files for strategic roadmap content

Triggers

  • Direct commands: /handoff, /handoff init
  • Refresh phrases: "update handoff", "session handoff", "refresh handoff", "save handoff"
  • Resume phrases: "cold-start primer", "resume checklist", "handoff to next session"
  • Stop hook: refresh silently when .docs/handoff.md already exists

Prerequisites

  • git available in the project root.
  • Project write access for .docs/handoff.md.
  • Optional project-local marker files:
    • .docs/handoff.last-validation containing the last verification command and result.
    • .docs/handoff.session containing extra bullets for changes, decisions, blockers, or refs.

Workflow

  1. Resolve mode.
    • /handoff init or invocation from an initialization workflow writes a greenfield stub when .docs/handoff.md is absent.
    • /handoff, refresh phrases, and Stop-hook invocation rewrite the full file.
  2. Read source of truth.
    • Load references/schema.md and preserve the schema headings exactly.
    • Load references/authority-boundaries.md and include the authority boundary line in every generated handoff.
  3. Detect repository state.
    • Run git rev-parse --show-toplevel, git branch --show-current, git rev-parse --short HEAD, and git status --porcelain.
    • Count modified, untracked, and staged paths from porcelain status.
    • Include the top 3-5 changed paths in the Status section.
  4. Detect validation state.
    • Prefer recent session context: commands actually run in the current agent session and their result.
    • If available, read .docs/handoff.last-validation.
    • If neither exists, write explicit unknowns: "not recorded this session" rather than inventing pass/fail.
  5. Gather session substance.
    • Summarize logical changes from current conversation, uncommitted paths, and commits since the branch base when relevant.
    • Record decisions with rationale in two sentences or fewer.
    • Record blockers and open questions as actionable unchecked items.
    • Include failed approaches when they would prevent wasted work in the next session.
  6. Write by full rewrite.
    • Replace .docs/handoff.md atomically; do not append.
    • Create .docs/ if absent.
    • Keep the file lowercase at .docs/handoff.md.
  7. Enforce 200 lines.
    • Count final lines after rendering.
    • If over 200, truncate ## What changed this session oldest-first.
    • Insert <!-- WARNING: handoff exceeded 200 lines; oldest session changes were truncated. -->.
    • If still over 200, compress decisions and blockers to the newest actionable items, preserving all required headings.

Supported Operations

OperationCommandBehavior
Refresh/handoffRewrites .docs/handoff.md from current repo and session state
Scaffold/handoff initCreates an empty schema stub when no handoff exists
Stop refreshStop hookRefreshes only when .docs/handoff.md already exists
Dry runuv run skills/handoff/scripts/handoff.py --dry-runPrints generated content without writing

Implementation Notes

The bundled script is the deterministic baseline used by the command and hook:

bash
uv run skills/handoff/scripts/handoff.py --project-root "$PWD"
uv run skills/handoff/scripts/handoff.py --project-root "$PWD" --init

Agents may improve populated bullets from live session context before writing, but must preserve:

  • Schema headings and order.
  • Authority boundary line.
  • Full-file rewrite semantics.
  • Hard 200-line cap.
  • Explicit unknowns instead of silent fallbacks.

Greenfield Scaffold

When invoked as /handoff init or under a project initialization flow and .docs/handoff.md does not exist, write the schema with empty placeholders and real git metadata. This documents the /init integration point without modifying any init script.

Output Format

Write exactly one file: .docs/handoff.md.

Required first lines:

markdown
# Handoff — <project name>

**Last touched:** <ISO 8601 timestamp with timezone> · **branch:** `<branch>` · **HEAD:** `<short-sha>` · **session:** <model id>

Troubleshooting

ProblemResolution
Not in a git repositoryFail clearly for refresh; scaffold may use branch and HEAD only during /handoff init
.docs/ missingCreate it before writing
Validation state unavailableWrite "not recorded this session"; do not claim tests passed
Generated file exceeds 200 linesTruncate oldest What changed bullets and add the warning comment
Existing handoff uses old casing or root pathDo not migrate automatically; future refreshes converge only when invoked in that project

Verification

  • uv run python scripts/validate_evals.py
  • uv run scripts/evaluate_package.py skills/handoff
  • uv run skills/handoff/scripts/handoff.py --project-root <repo> --dry-run
  • uv run skills/handoff/scripts/handoff.py --project-root <repo> --init --output <tmp-file>

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Handoff AI skill do?

Produces and refreshes `.docs/handoff.md`, a 200-line session-continuity runbook for coding agents. Captures git state, working-tree changes, last validation commands, session decisions, blockers, resume checklist, references, and greenfield scaffolds. Triggers on: "update handoff", "session handoff", "refresh handoff", "/handoff", "/handoff init", "resume checklist", "cold-start primer". Use this skill when ending, pausing, resuming, or transferring in-flight project work across agent sessions.

Why use Handoff on TypingMind?

Because you install it once and use it with any model. Handoff is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Handoff in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Mathews-Tom/armory/tree/main/skills/handoff. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Handoff?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Handoff?

As many as you like. As long as a model supports skills, you can use Handoff with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Handoff AI skill free?

Yes. It is published on GitHub by Mathews-Tom under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇