Scheduled Auto-Archive: An AI Agent That Archives Old Files Before You Notice Them

Published on November 3, 2026

Quick answer: FinalPlace is a Windows rule engine for local file organization, and two of its building blocks make scheduled auto archive possible: a file-age condition (based on a file’s calendar age or its last-modified time) and archive-oriented actions such as move and compress. Once a rule is enabled, the engine enforces it quietly in the background, so files that sit past their deadline archive themselves. Pair the CLI with an AI agent and you can get there without filling in a settings form: describe what “old” means and where things should go, the agent assembles the rule, previews it with zero changes to your files, and executes after your confirmation. Everything runs completely local and offline — your files never leave the machine.

The folder that ages faster than you tidy it

Downloads is a time capsule. The installer you needed once, the receipt from a hotel stay two quarters ago, the PDF somebody emailed “just in case” — none of it is trash, so none of it gets deleted, and the folder grows until scrolling it feels like archaeology. The intent to fix it is real. What fails is the execution, because archiving by hand means deciding file by file what counts as old, where each thing belongs, and whether this folder needs zipping. That is an afternoon of judgment calls, and the mess outlives another weekend.

Industry surveys consistently estimate that office workers lose one to two hours per week locating files they have already saved. Search can shave that down, but search treats every file as equally current. The real problem is that nothing ever leaves: outdated documents sit at the same priority as the ones you opened this morning. The fix is to make age itself the trigger — to automatically archive outdated documents on a schedule, so “old” stops being a judgment you make and becomes a condition a rule engine evaluates.

What auto archive by age looks like in FinalPlace

FinalPlace’s rule engine offers eight condition types, and two of them are about time: file age and last-modified time. “Older than 30 days” and “untouched for 90 days” are each a single condition a rule can be built on. On the action side, twelve action types include move, copy, rename, and compress — with compress turning an archive folder from “the same files, somewhere else” into a genuinely smaller archive. Four conflict-handling strategies decide what happens when the destination already holds a file with the same name, so a collision never becomes a silent overwrite.

Once a rule is enabled, it executes quietly in the background. There is no batch session to remember and no window to babysit — old files simply stop accumulating where you work. Industry surveys consistently estimate that well over half of the files stored on typical business computers are never opened again after the week they arrive. That silent majority is exactly the population an age-based rule should be sweeping, and no amount of good intentions sweeps it as reliably as a condition that checks the clock on every pass.

Let an AI agent build the archiving rules

This is the fifth article in our series on running FinalPlace with an AI agent; the series overview explains how the CLI and the agent skills package fit together. The short version: the agent reads the engine’s vocabulary — the condition and action templates — and translates plain-English archiving wishes into well-formed rules.

The executable lives at "C:\Program Files\FinalPlace\FinalPlace.exe", and the agent discovers what the engine supports through listing commands such as:

"C:\Program Files\FinalPlace\FinalPlace.exe" condition-templates list
"C:\Program Files\FinalPlace\FinalPlace.exe" action-templates list

From there, you describe the outcome. Something like: “Everything in my Downloads folder older than 30 days should move into a year-based archive folder, and anything older than 90 days should be compressed.” The agent drafts the rule, then runs a preview that changes nothing — a dry run reporting exactly which files would go where. If the plan looks right, you confirm and the agent executes. If you have not connected an assistant yet, the installation and setup guide walks through the agent skills setup step by step.

Four age-based recipes to paste to your agent

Each recipe is a request you can paste to your agent, with the reasoning behind it. Treat every path as an example — the agent adapts them to your machine.

The 30-day Downloads sweep

Request: “Files in Downloads older than 30 days move to my archive folder, grouped by year.”

A file-age condition paired with a move action. This keeps Downloads a working set of things you actually asked for recently, while everything else graduates to the archive without a decision from you.

Quarterly document archiving

Request: “Documents I have not touched for 90 days go to the archive, grouped by quarter, and old invoices go to a Finance archive instead.”

The last-modified-time condition catches documents a quarter past their use, and the target path can carry date placeholders so the archive organizes itself as it grows. The agent can also combine the age condition with contentKeywordMatch — the condition type that reads inside documents — so an aging invoice routes to Finance while an aging contract routes to Legal, decided by content rather than filename.

Shelve finished projects with compression

Request: “When nothing in a project folder has changed for 30 days, compress the folder and move the archive to Deliverables.”

Here the last-modified time acts as a “done” signal: when a whole folder has been quiet for a month, the project is presumably finished, and the compress action packs it into a single archive file. The original folder stops taking up attention — and the compressed copy takes up less disk.

A gentle start with copy

Request: “Copy files older than 60 days into the archive, but leave the originals where they are.”

The copy action doubles as a trial mode. Run it for a month, watch the archive take shape, and confirm nothing you needed went missing; then ask the agent to switch the same rule from copy to move. For a wider set of condition and action combinations, the rules cookbook from the previous article collects seven recipes, several of which pair nicely with an age condition.

Quiet, local, and reversible

Everything described above runs completely local and offline. The rule engine moves and compresses files on your own disk, between your own folders — no cloud upload happens at any step, so archived contracts, invoices, and personal documents stay on the machine. Industry surveys consistently estimate that a majority of office workers feel uneasy about uploading personal or company documents to online services, and local processing sidesteps the question entirely.

Agent-built rules are not permanent experiments either. Every run is logged, so you can review what a rule did and roll back a run that did not go as intended. The recommended rhythm stays the same throughout the series: preview with zero changes, read what the agent shows you, confirm deliberately, and keep the log as your undo history.

Pricing

Licensing on the English site has four options: 3 months for $10, 1 year for $28, 3 years for $58, and a lifetime buyout for $88. The 3-month plan is long enough to run a full quarterly archiving cycle — sweep, archive, compress, review — before you commit to a longer horizon.

FAQ

How to archive a file in Windows automatically?

Point a rule at it. In FinalPlace, an archiving rule pairs a time condition (file age or last-modified time) with an action such as move or compress; once the rule is enabled, the engine applies it in the background without further attention. With the agent flow, you can get such a rule by describing it in one sentence and confirming the zero-change preview.

Do the archive rules run on a schedule without me?

Yes, in the sense that matters: once a rule is enabled, the engine enforces it quietly during normal operation — there is no monthly chore to remember and no button to press. The agent’s role is building and verifying the rule; the ongoing enforcement belongs to the engine.

What happens if the archive already contains a file with the same name?

The rule engine provides four conflict-handling strategies, and one is selected when the rule is created. Ask your agent for “never overwrite” behavior in your request, and it applies the matching strategy — for example keeping both files with a renamed copy — so collisions resolve without data loss.

Do my archived files leave my computer?

No. FinalPlace runs completely local and offline: files are matched, moved, and compressed on your own disk, and nothing is uploaded anywhere.