A workflow for handing work from one coding session to the next when the current context is full. It records the project state, next actions, decisions, and a daily log, then queues conversation compaction and continuation from the handoff.

Use it for the supported rollover commands or when asked to compact and continue a captain session. It helps preserve open pull requests, crew work, and other project details across sessions.

It reduces the chance of losing active work or important decisions during a long session. It also ensures compaction is the final action rather than interrupting unfinished work.

rollover — Skill for Claude Code by tu11aa/squadrant

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.

Any agent

npx skills add tu11aa/squadrant --skill rollover

Clone the repo

git clone --depth 1 https://github.com/tu11aa/squadrant

Made for: Claude Code.

Or install squadrant, the plugin that ships this one along with the rest of its 27 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/tu11aa/squadrant/rollover/github.svg)](https://agentmods.dev/skills/tu11aa/squadrant/rollover)

Your own site

<a href="https://agentmods.dev/skills/tu11aa/squadrant/rollover"><img src="https://agentmods.dev/badge/skills/tu11aa/squadrant/rollover/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/tu11aa/squadrant/rollover"><img src="https://agentmods.dev/badge/skills/tu11aa/squadrant/rollover.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 68 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 515 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00068	$0.00515
Opus 5.5	$0.00027	$0.00206
Sonnet 5.5	$0.00014	$0.00103
Haiku 4.5	$0.00007	$0.00052

Measured yesterday against content hash b2a7f64117ca, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

rollover 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.

plugin/skills/rollover/SKILL.md · 22 lines

What it actually says

rollover — handoff → daily log → auto /compact → continue

Run once; the operator walks away. Claude Code cannot compact itself, so the daemon types /compact into this pane after your turn ends, and the SessionStart hook hands your handoff back afterwards.

Steps (in order)

Write a specific handoff with ~/.config/squadrant/scripts/write-handoff.sh "{project}" '{…}' (fields and the consolidation rule — read the current handoff, drop verified-done, merge duplicates, keep all open items and decisions — in captain-ops → Session Shutdown). Be concrete: currentState says what is in flight and where; nextSteps is the first actions to take, with PR/issue/crew ids; decisions lists what must not be revisited. Do not run read-handoff.sh — it would archive the file.
Write today's daily log with the squadrant:daily-log skill.
Queue the compact as your LAST action:
```
squadrant captain compact {project}
```
Optional --focus "<text>" replaces the default ("Keep the active task, open PRs/crews and decisions; the full state is in the handoff"). It returns immediately.
End the turn with one short line (e.g. "Handoff + log written; compacting."). Do nothing else — a running turn delays the compact, and the daemon never types over a busy pane or an open prompt (it gives up after 10 min and logs it).

After the compact, the handoff arrives as session context with "continue from nextSteps" — resume from it without asking the operator. /compact is a local command that starts no turn, so right after typing it the daemon queues one captain message ("Compact finished — continue from the handoff nextSteps.") over the peer channel; Claude Code delivers it when the compact ends and it starts your turn. Only compacts squadrant queued get this; a /compact typed by hand does not.

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.

yesterday Changed b2a7f64117ca
15d ago Changed c4f6ee08a8c2
17d ago First seen · 22 lines · 68 tokens per session scan A 184ac4b3959f

Subscribe to this mod's changes

rollover is a skill published in the GitHub repository tu11aa/squadrant (2 stars, last pushed yesterday), licensed MIT. It adds 68 tokens to every session and 515 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-20.

Productivity Project management

rollover next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
rollovertu11aa/squadrant	Skill	2	yesterday	MIT	A	68	Claude Code
orbit-notionnexu-io/open-design	Skill	106k	yesterday	Apache-2.0	A	117	Claude Code, Codex
agentmailpaperclipai/paperclip	Skill	111k	yesterday	MIT	A	30	Claude Code, Codex
Cortexdanielmiessler/LifeOS	Skill	20k	yesterday	MIT	A	184	Claude Code
rowboat-spacesrowboatlabs/rowboat	Skill	18k	yesterday	Apache-2.0	A	60	Claude Code, Codex
feishuHmbown/CodeWhale	Skill	41k	5d ago	MIT	A	33	Claude Code, Codex
pinchtab-mcppinchtab/pinchtab	Skill	11k	10d ago	MIT	A	52	Claude Code, Codex