handover-check

handover-check is a command for coding agents from FlorinI/Florians-Claude-Code-Kit. It costs 29 tokens per session (3,323 once invoked), scanned A, original, MIT.

A command that reads a session status snapshot and explains whether continuing or handing work over is the better choice.

In plain words
What is it for?
Use it to assess session cost versus answer quality and get a plain-language recommendation about handing work to a new session.
Why use it?
It translates usage and quality signals into a practical decision about when the current coding-agent session should end.

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/florini/florians-claude-code-kit/handover-check
Clone the repo
git clone --depth 1 https://github.com/FlorinI/Florians-Claude-Code-Kit

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 handover-check

README.md
[![agentmods](https://agentmods.dev/badge/commands/florini/florians-claude-code-kit/handover-check.svg)](https://agentmods.dev/commands/florini/florians-claude-code-kit/handover-check)
Your own site
<a href="https://agentmods.dev/commands/florini/florians-claude-code-kit/handover-check"><img src="https://agentmods.dev/badge/commands/florini/florians-claude-code-kit/handover-check.svg" alt="Measured on agentmods" height="20"></a>
Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,323 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 $0.00029 $0.03323
Opus 5 $0.00015 $0.01662
Sonnet 5 $0.00006 $0.00665
Haiku 4.5 $0.00003 $0.00332

Measured 4d ago against content hash 52f6d2ac8b09, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

handover-check 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 4d 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.

home/commands/handover-check.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/handover-check

Explain the current session's status line to the user in plain, practical terms — the read a smart person who doesn't live in Opus session-management would want.

The heavy lifting is deterministic: a Layer-1 script (handover-facts.mjs) resolves every classification/verdict and pre-formats every number, so the isolated subagent only has to compose — weave the facts into insightful prose. This keeps the check off the very context it measures, kills the band-misread and backtick-mangle classes (the model can't fumble what it's handed), and front-loads every tool call so the report is one uninterrupted reply (Claude Code collapses/fragments tool output — nothing must land mid-prose).

How to run

Two phases — all data-gathering first, then one report.

Phase 1 — gather (emit NO prose yet). Order matters: the fact sheet runs FIRST and freezes the exact snapshot, then the two charts read that frozen copy (--frozen) so all three see the SAME session — even if a concurrent same-project session clobbers the live sidecar between reads (the within-report session-split bug).

  1. Run the fact sheet alone first — it writes the freeze. ($HOME expands in both PowerShell and bash; Node accepts the forward slashes on Windows too. On Florian's Windows boxes, use the PowerShell tool per the project convention.)

    node "$HOME/.claude/handover-facts.mjs"
    

    If it prints MISSING, tell the user to let the status line render once (press Enter on an empty prompt) and re-run — stop here. If its first line is FOREIGN, this session has no status-line snapshot of its own and the resolver fell back to another session's sidecar (the following lines give this session vs the snapshot's owner — each named in words with its id fragment in brackets when that session has a name, id fragment alone otherwise; relay them as given). This is the signature of a session that never renders a status line — the Claude Code desktop app is the common case. Do NOT report the numbers: they belong to a different session and would be meaningless here. Tell the user plainly that handover-check can't read this session (no live status line — e.g. the desktop app) and that it works from a terminal session where the status line renders — then stop here. If its first line is SCHEMA_MISMATCH, the snapshot was written by a status line of a different build than this reader, so its shape is one this reader does not know. The following lines give the schema version this reader expects, the version the snapshot carries, and the one instruction that clears it — relay all of them as given. Do NOT report any numbers, and do NOT run the charts or the subagent: there is no reading here, only the fix. Then stop here.

  2. Then, in one message, run the two charts with --frozen AND spawn the subagent (they're independent):

Read the full file on GitHub · 78 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. 4d ago First seen · 78 lines · 29 tokens per session scan A 52f6d2ac8b09

Subscribe to this mod's changes

handover-check is a command published in the GitHub repository FlorinI/Florians-Claude-Code-Kit (5 stars, last pushed 6d ago), licensed MIT. It adds 29 tokens to every session and 3,323 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.