---
name: zerto-rewind
description: Before changing a VM, find its Zerto VPGs and insert a tagged checkpoint. Recover files from that tag with FLR after a human confirms. Use whenever an agent will mutate a guest that might be protected by Zerto.
---

# Zerto rewind

Zerto already journals the VM. This skill makes the agent use that journal. Git does not have the guest file. Official ZVM MCP does not insert tagged checkpoints.

You talk to **one** MCP: `zerto_rewind_mcp`. Do not also require official ZVM MCP.

## Loop (mandatory)

Before **every** tool call that might touch a guest:

1. Take the hostname / VM name / Zerto `vmIdentifier` from the tool args.
2. Call `zerto_check_tool(server, tool, vm)`. Act on the verdict:

| verdict | what you do |
|---|---|
| `read_only` | Run the tool. No checkpoint. |
| `mutating` | Guard first. Do not ask — it is already known to change the guest. |
| `unknown` | **Ask the human.** Do not assume it is safe, and do not silently guard. |

3. For `mutating`, or for `unknown` where the human said yes: call
   `zerto_guard_before_mutate` with `change_id` and `action`.
4. If `ok` is not true: **stop**. Do not mutate.
5. Then run the call.

`unknown` is the normal case, not an edge case. The catalogs are short and the
world of tools is not, so most tools are unclassified. Unknown means *nobody has
said this is read-only* — it does not mean safe. Put the decision to the human:

> `winrm/run_ps` is not a known read-only command. It may change `web01`.
> Insert a Zerto tagged checkpoint first so this is reversible?

If they say yes, guard, then run. If they say no, run it and tell them plainly
that it is not reversible through Zerto. If they want it remembered, call
`zerto_add_mutating_tool` (server, tool, `vm_arg`) so it is guarded
automatically next time instead of asking again.

## find_protection outcomes

| outcome | what you do |
|---|---|
| none | Unprotected or unknown. Refuse the change. Say Zerto cannot rewind this. |
| ambiguous | Two or more VMs matched. Ask for a `vmIdentifier`. Do not guess. |
| ok, no taggable VPG | Syncing or not Protecting. Refuse. A resync deletes checkpoints. |
| ok, taggable VPGs | Tag **every** protecting VPG with the same tag. Wait until listed (the tool blocks). |

A VM can be in up to three VPGs (local backup + remote DR is common). Tag all of them.

## Recover

Human must confirm. Pass `confirmed=true` only after they say yes.

- Bad config / dropped file: `zerto_recover_file` from **that tag**. Pass the guest path
  (`/home/j/app.yaml` or `C:\Users\j\app.conf`); the server maps it into the FLR
  namespace. **Locally replicated VPGs only** -- FLR happens at the recovery site, so a
  cloud-replicated VPG must be recovered from that ZCA. The tool refuses and names the site.
- Inspect a whole VM: `zerto_offsite_clone` or `zerto_start_failover_test`.
- Never Failover Live. Never Move. Those are DR, not rewind.

`zerto_recover_file` unmounts its own FLR session and reports the result in
`unmount`. If `unmount.ok` is false, or a previous recovery died mid-flight,
the mount is still up: FLR cannot run during clone, test, live failover or EJC,
so a stuck session blocks the next recovery. Find it with
`zerto_list_flr_sessions` and clear it with `zerto_end_flr_session`.

## Facts that bite

- A tagged checkpoint is crash-consistent, not app-quiesced.
- Tagged checkpoints are not supported when the **protected** site is Azure or AWS. Talk to the vSphere protected ZVM.
- 10.9 FLR Operator RBAC fails; Administrator is the documented workaround.
- FLR cannot run during clone, test, live failover, or EJC.
- Linux FLR: files >1.5GB are a bad idea; some characters in names are refused.

## Tag

The checkpoint name is the only field the Zerto API takes, so it carries the
whole story:

```
ai:<agent> | <action> | vm=<vm> | change=<change-id> | <utc>
ai:claude | edit /etc/nginx/nginx.conf | vm=web01 | change=chg-412 | 20260921T150405Z
```

Always pass `action`: a plain description of the change you are about to make.
An operator scrolling the journal in the Zerto UI should be able to tell which
agent inserted the checkpoint and why, without reading your transcript.

Same string on every VPG for that call.