handover

A command for recording the current state of unfinished work so it can be resumed later.

In plain words
What is it for?
Use it when work is genuinely blocked, paused mid-task, or being handed to another session, with notes stored at an appropriate level.
Why use it?
It preserves evidence-based details about completed work, decisions, and the next action when a session is paused or moved to another environment.

Command for Claude Code

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/littlebearapps/untether/handover
Clone the repo
git clone --depth 1 https://github.com/littlebearapps/untether

Made for: Claude Code.

Per session 114 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,549 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 $0.00114 $0.01549
Opus 5 $0.00057 $0.00775
Sonnet 5 $0.00023 $0.00310
Haiku 4.5 $0.00011 $0.00155

Measured 2d ago against content hash 08b9cf4d362a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

handover 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.

.claude/commands/handover.md · 131 lines

How it starts

The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are handling /handover. /handover captures resumable stop-state so a paused/blocked session can be picked up cleanly later — critical for Telegram sessions that may resume on a different host or after a gap.

User input: $ARGUMENTS

Untether adaptations (read first)

Load .claude/rules/workflow-commands.md (routing + cross-cutting rules). Key points:

  • Derive from persisted state, never chat memory. complete[], decisions[], and next_action come from evidencegit (branch, HEAD, diff, stashes), test runs, session_quarantine.json, active_progress.json, and logs — not from what you "remember" doing. Chat memory is unreliable across a resume; the repo is ground truth.
  • Default DOWN between levels. A routine session end is H0 (no handover). Only escalate when work is genuinely paused mid-flight with an unobvious next action. Never manufacture an H2/H3 for a clean stopping point.
  • No auditor continuity validator — Untether uses a plain frontmatter template + a git-based fingerprint. No manifest CLI.
  • Authority: writes the handover doc only. Never code, never a PR, never a release action. H2 is gitignored; H3 is committed (docs only).
  • Redaction. Scrub tokens/keys/env/chat-content/fleet identifiers from any log/state evidence before it lands in a committed (H3) doc.

Levels (default DOWN)

Level When Where
H0 routine end, nothing paused nothing written (the common, correct case)
H1 a one-line "next time, do X" an inline note in the final message
H2 paused mid-work, private/local incoming/handovers/<YYYY-MM-DD>-<slug>.md (gitignored)
H3 a durable, shareable handover worth committing docs/handovers/<YYYY-MM-DD>-<slug>.md (committed)

--level HN forces a level; otherwise auto-pick, biasing DOWN.

Flow

V-1. Decide the level

Is work genuinely paused with a non-obvious next action? If noH0, say so, stop. If a single line suffices → H1. Escalate to H2/H3 only for real mid-flight interruption. When unsure, go one level DOWN.

Read the full file on GitHub · 131 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 · 131 lines · 114 tokens per session scan A 08b9cf4d362a

Subscribe to this mod's changes

handover is a command published in the GitHub repository littlebearapps/untether (66 stars, last pushed 3d ago), licensed MIT. It adds 114 tokens to every session and 1,549 once invoked, about $0.0006 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.