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.
npx skills add Vinix24/vnx-orchestration --skill rotategit clone --depth 1 https://github.com/Vinix24/vnx-orchestrationWrote 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.
[](https://agentmods.dev/skills/vinix24/vnx-orchestration/rotate)<a href="https://agentmods.dev/skills/vinix24/vnx-orchestration/rotate"><img src="https://agentmods.dev/badge/skills/vinix24/vnx-orchestration/rotate/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.
<a href="https://agentmods.dev/skills/vinix24/vnx-orchestration/rotate"><img src="https://agentmods.dev/badge/skills/vinix24/vnx-orchestration/rotate.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00069 | $0.02034 |
| Opus 5.5 | $0.00028 | $0.00814 |
| Sonnet 5.5 | $0.00014 | $0.00407 |
| Haiku 4.5 | $0.00007 | $0.00203 |
Grade A, and why
rotate 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.
How it starts
The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Rotate: hand this session over to a fresh one
A long session gets worse before it compacts. A handover is a short file that tells a
fresh session everything it needs to continue. Then /clear starts that fresh session:
inside tmux the plugin types it into this session's own pane once the session is idle,
elsewhere the user types it. The plugin gives the handover to the new session.
The first word of the arguments picks the action: write (also with no arguments),
release or resume. For any other first word the plugin puts its usage line in your
context: show that line to the user and do nothing else.
write
-
Find the path. It is in your context: the guard names it in every message about the context size, and when the user typed this command a note gives it. It looks like
<project>/.claude/rotation/handover-<session id>.md. Never write another session's handover. -
Bring the work to a resting point first: finish the edit you are in, or stop cleanly. The handover describes the state at the moment you write it.
-
Write the file with the Write tool. Not with a shell command: the check runs on Write and Edit. Use these sections, in this order:
# Handover ## Goal What the user wants, in their terms, and how they will judge it done. ## Current state What is done and verified, the decisions taken and why, what was tried and failed. ## Running Background shells, subagents, servers or watchers that still run, with how to find them. Write "none" when nothing runs. ## Next steps 1. The first thing the fresh session does: a command, a file, a question to the user. 2. ... ## Files touched Each path with one line on what changed. ## Open risks What is unverified, what can break, what the user still has to decide.Rules for the text:
- Write for a reader who has seen none of this conversation: full paths, exact names, the exact command and its last result. No "as discussed".
- The first line under
## Next stepsmust be a step numbered1.. - At most 8,000 characters, or the limit named in your context.
-
Read the answer of the check. Right after the write the plugin either says the handover is valid and marked ready, or names exactly what is missing. Fix it with another Write or Edit until it is valid.
-
Make the valid write your last action. Do not call any other tool afterwards, not even a todo update: any call other than Read, Glob or Grep, and any new prompt, marks the handover outdated, and an outdated handover is never given to a fresh session. End your turn with one short line saying the handover is saved.
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.
- 4d ago Changed · +18 lines e1dc92ae72e3
- 5d ago First seen · 124 lines · 69 tokens per session scan A 11b99e767a21
rotate is a skill published in the GitHub repository Vinix24/vnx-orchestration (62 stars, last pushed today), licensed MIT. It adds 69 tokens to every session and 2,034 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 2026-10-18.
Other skills, from other repositories
skill-iterative-loop
Run tasks in a loop until goals are met — use for iterative refinement, polling, or convergence.
storage-retention
Inspect machine storage, clean up files, and configure automatic thread archive or deletion.
prompt-library
Search prompt history and manage starred reusable prompts in BB.
log-work
Log the work segment that you did after the last work log until now in a log file at WORKLOG/YYYYMMDD.md where YYYYMMDD is based on today's date. The intent of this worklog is not to be exhaustive and super detailed. It is meant more as a very concise description plus, importantly, REFERENCES to more detailed…
oma-recap
Summarize AI conversation histories for a specified date or period. Use for daily work recaps and cross-tool activity summaries.
sys-configure
Configure Claude Octopus — redirects to /octo:setup interactive wizard.