What is Pluckeye?

Pluckeye is an internet filter for people who want to limit their own internet access. It runs on Windows, macOS, Linux and Android. What makes it different from every other filter is that it has no password — and it does not need one.

The one-paragraph version

You install it. You write rules saying what is blocked and what is allowed. You are the administrator, so you can change those rules whenever you like. Then you set a delay — say four hours — and from that moment on, any change that gives you more access sits in a queue until the delay has passed. Changes that give you less access happen instantly. That is the entire idea, and everything else is detail.

Why a delay beats a password

Filters that use a password have to hide the password from you. Either someone else holds it, which means bothering them every time something legitimate is blocked, or you hold it and it stops working the first time you really want through.

A delay makes a different bet: that the version of you at eleven at night and the version four hours later want different things, and the second one should get to decide. You never have to ask anyone’s permission. You just cannot have it now.

How the delay actually behaves

More access is delayed. Allowing a site, unblocking a media type, removing a block rule — all of these wait.
Less access is immediate. Blocking something new, or removing an allow rule you regret, applies the moment you press enter. You are never stuck waiting to be stricter.
Lowering the delay is itself more access. Ask for zero while sitting on four hours and you wait four hours. This is what stops the delay being cosmetic.
Uninstalling needs a delay of zero. Which, per the point above, is four hours away.

pluck delay        # what is it currently?
pluck delay 4h     # four hours
pluck delay 45m    # forty-five minutes
pluck delay 1d     # a day
 
# a different delay for one kind of change:
pluck delay 24h "+ allow everything"

Start smaller than feels right. An hour that you actually keep is worth more than a week you panic out of on day two — and because lowering it is slow, a delay you regret is genuinely annoying to undo. The manual’s delay page has the full behaviour.

Installing it

Normal installer, and no account needed to get going. Pick your platform:

Once it is installed, the first-steps page has a five-minute demo — allow an image, set a thirty-second delay, watch the delay work, delete the rule. Do it before you configure anything you care about. It teaches the mental model faster than any amount of reading.

Choosing a base configuration

On first run you pick a starter pack. Both are just a configuration you import and then build on, so this is a starting point rather than a commitment.

Wimpy

Blocks only the obviously bad stuff. The rest of the web behaves normally. If you came here because of YouTube specifically, this is the one — you will add the YouTube rules yourself and nothing else breaks.

Classic

Blocks all images and videos on all websites by default; you allow back what you need. Much stricter, and much quieter. Choose it if the problem is bigger than one site, and expect to spend the first week allowing things.

There is also whitelist-only mode for people who want nothing at all except a named list — and, at the other end, focus for temporarily locking down while you work.

How rules are written

A rule is an action plus a context, optionally limited to a time window. The actions are allow, block, blot (strip images from pages mentioning a word) and flee (block any page that even links to something), each with an un… form that cancels it.

pluck + block image/                     # all images, everywhere
pluck + allow example.com                # a host, and its subdomains
pluck + allow https://example.com/docs   # a URL, and everything under it
pluck + block program qbittorrent        # a program
pluck + block youtube.com video/         # one media type on one host
pluck + when MTWRF9-17 block reddit.com  # only during the working week

Two behaviours do most of the work: host rules automatically cover subdomains, and URL rules automatically cover every URL underneath them. The second is what makes whitelisting a YouTube channel possible at all.

Useful reference: rules, commands, media types, when.

Commands you will actually use

pluck + <rule>      # add a rule
pluck - <rule>      # remove one
pluck delay         # print the delay
pluck verdicts      # what did it just block, and why
pluck find reddit   # which of my rules mention this?
pluck export        # print the whole configuration
pluck abort         # cancel changes still waiting out the delay

pluck verdicts is the one to remember. Almost every “why is this broken” moment is answered by it.

Common questions

What is the user site at u.pluckeye.net?

An account-backed companion to the program. It is what makes syncing configurations between devices work, along with approvers, inspectors and imported configurations, and it is required to customise the Android version. Nothing described on this page needs it — a single machine driven from the console does not touch it at all.

Can I just uninstall it when I want back in?

Only when the delay is zero. Since lowering the delay has to survive the delay itself, uninstalling is exactly as slow as you decided in advance it should be.

Does it only filter the browser?

No. Pluckeye filters at the network level, so it covers other programs too. It is a network filter rather than an app locker: blocking a program stops it reaching the internet, it does not stop it opening.

What is the difference between Pluckeye and Plucky?

They are the same thing. Pluckeye is the project and the website; Plucky is what the program is called, and it is the name the manual mostly uses. The command you type is pluck.

Does Pluckeye decide what is good and bad for me?

Barely. It ships with starter configurations, but the design leans towards giving you the controls rather than a curated blocklist. Almost everything is a rule you wrote.

Then what?

The most common reason people install Pluckeye is YouTube, and it is also the site it handles least well out of the box, because YouTube’s URLs do not mention which channel a video belongs to. That is the gap this site fills.