Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add motlin/claude-code-plugins/plugin install herdr-rebootWrote 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/motlin/claude-code-plugins/restore)<a href="https://agentmods.dev/skills/motlin/claude-code-plugins/restore"><img src="https://agentmods.dev/badge/skills/motlin/claude-code-plugins/restore/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/motlin/claude-code-plugins/restore"><img src="https://agentmods.dev/badge/skills/motlin/claude-code-plugins/restore.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 | $0.00054 | $0.00782 |
| Opus 5.5 | $0.00022 | $0.00313 |
| Sonnet 5.5 | $0.00011 | $0.00156 |
| Haiku 4.5 | $0.00005 | $0.00078 |
Grade A, and why
restore 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 22d 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.
What it actually says
Restore Herdr Workspaces
Rebuild the session from the snapshot written by /herdr-reboot:snapshot: each workspace and tab with its label, panes split at the captured direction and ratio, then the resume command fired into each pane.
Resolve <plugin-root>: ${CLAUDE_PLUGIN_ROOT} in Claude Code; in Codex, the plugin root containing this loaded skills/restore/SKILL.md, even when installed outside the repository.
Preview the plan and show it to the user:
python3 "<plugin-root>/scripts/restore.py" .llm/resume-after-reboot-state.json
The preview prints workspaces, tabs, split right 0.6 lines, and the pane each slot lands in. Once the user confirms, fire it:
python3 "<plugin-root>/scripts/restore.py" .llm/resume-after-reboot-state.json --go
--limit N rebuilds only the first N workspaces; --skip 3,7 leaves slots out (a split left with one pane collapses, and a tab left with none is not created). Agent launches are spaced 0.4s apart (--delay).
Only schema resume-after-reboot/v2 is accepted. A flat resume-after-reboot/v1 snapshot is rejected rather than half-restored.
Tell the user before firing:
commandpanes fire by default, since bringing back dev servers likejust devis most of the point.--no-commandsleaves dev servers (restore_default: false) alone when one outlived the reboot and holds its port. Read-only viewers (git log,less FILE,htop,tig) fire either way, from the top, without scroll position, search, or selection.- herdr may already have brought the session back, agents included, leaving only command panes at a prompt. In the preview,
(adopt …)on a workspace line and(live …)on a tab line mean restore reuses what exists. Each captured tab pairs with the live tab of the same label and split structure; its idle panes are fired into in place, and a pane already running its agent printsLIVE. Only a captured tab with no live counterpart (renamed or re-laid-out since the snapshot) is created fresh. - Nothing is typed into a pane that is not an idle shell.
herdr pane runtypes into whatever the pane holds, so firing into a resumed agent would submit the command as a prompt. Panes not confirmed at a shell prompt with no live agent are skipped with a printed reason, so re-running is safe. - A live workspace with the same label and working directory is adopted, not duplicated.
- Focus comes back last: each workspace's active tab, then the focused workspace. In-tab focus is restored by creating the focused pane focused.
- Verify codex panes whose note says
herdr reported no session; their ids are best guesses from matching the working directory against rollout files. If one opens the wrong conversation, runcodex resumein that pane and use the picker. claude-rcpanes are Remote Control servers, fired like any agent pane.--continueerrors if nothing was recorded for the directory in roughly the last 4 hours, which is expected after any reboot you did not come straight back from. The pane is left at a prompt; start a fresh server by hand if wanted.
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.
- 22d ago Changed · -31 lines a620a0795e05
- 1mo ago Changed · +3 lines · -16 tokens per session 0cbc677b3093
- 1mo ago Changed · +6 lines 4a229de49f57
- 1mo ago First seen · 59 lines · 70 tokens per session scan A 60eae2a6bc06
restore is a skill published in the GitHub repository motlin/claude-code-plugins (20 stars, last pushed yesterday), licensed Apache-2.0. It adds 54 tokens to every session and 782 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
agentmail
Use your assigned AgentMail inbox to read email tasks, explicitly send or reply, and check delivery. Provided automatically by your inbox assignment.
rowboat-spaces
How to work as a member of Rowboat Spaces, the team chat where people and agents work together. Load it once per session, before acting on any request that reaches you from Rowboat; it covers your setup, mentions and hand-offs, the tools, and files.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…