Clear your Instagram like history without looking like a script. Emulates a real Android device, mixes in ordinary browsing between actions, and works oldest to newest so the deep history goes first.
English · Türkçe · Español · 简体中文 · Русский
Install · Features · Usage · Tutorial · Changelog
- Behaves like a person, not a loop - most bots produce a perfectly even click, click, click. This one scrolls the home feed, watches reels for random lengths and opens DMs between unlikes, because the gaps are what make the pattern look human.
- Real device emulation - runs as a current Android flagship, with matching screen dimensions, touch behaviour and user agent, rather than a desktop browser pretending.
- Oldest first - the like history is sorted chronologically and cleaned from the far end, so the posts you forgot about years ago go before the ones from last week.
- Randomised pacing, not fixed timers - every delay is drawn from a range. A constant interval is the single easiest signal to detect.
- Hourly ceilings - a hard cap on how much happens in an hour, so a long session cannot quietly turn into a burst.
- Automates the Sort and Filter interface - Instagram's own filtering is driven directly instead of being worked around.
- Stays logged in - the session is saved so you are not re-authenticating on every run.
navigator.webdrivermitigation - the usual automation giveaway is removed before the page ever sees it.- Readable logs - each cycle prints its plan and its progress, so you can watch what it intends to do before it does it.
| Python | 3.8 or newer |
| Playwright | with the Chromium browser installed |
| Account | Your own Instagram account |
The tool ships as an archive attached to the release, in an English and a Turkish build.
- Open the latest release.
- Download
Munliker-ENG.rarorMunliker-TR.rar. - Extract it somewhere you can work.
pip install playwright
playwright install chromiumCloning this repository does not give you the program. The repository holds the documentation; the code is distributed through the release archives above.
python main.pyOn the first run you sign in once and the session is stored so later runs skip it. The console then prints its plan for the cycle: how many posts it intends to unlike, the delays it will use and the filler actions it will mix in. Watch that first cycle before you leave it alone.
Slow is the point. This is not a race. Instagram restricts accounts on burst behaviour, and the defaults are chosen to stay well under that. Raising the limits is the fastest way to lose the account you were trying to tidy.
Everything lives in config.py. The settings that matter most:
| Group | What it controls |
|---|---|
| Speed and limits | Unlike delays, the hourly ceiling, and how long a cycle runs |
| Device emulation | The device profile the browser presents |
| Filler actions | Whether to scroll the feed, watch reels and open DMs, and how often |
| Session | Where the login session is stored |
Each field, its accepted range and what happens when you push it too far is documented in the tutorial.
Keep
cookies.jsonto yourself. It holds your logged-in Instagram session. Anyone who has that file is signed in as you. It is excluded by.gitignore, and it must never go into a repository, an issue or a screenshot.
- Tutorial - the architecture, every module and every setting
- Changelog - what changed in each release
- Contributing - how to send a change
- Security - how to report a vulnerability privately
Can this get my account restricted?
It can. Anything that acts on your account faster or more regularly than you would is a risk, and this tool exists precisely because the naive version of it gets caught. The defences here are randomised pacing, filler actions and hourly ceilings, and they lower the risk rather than remove it. Leave the defaults alone, run one session at a time, and stop if Instagram shows you a warning.
Why is there no code in this repository?
The program is distributed as an archive attached to each release rather than as files in the repository tree. Download it from the releases page.
Why unlike oldest first?
Because the old end of the history is the part you have actually forgotten, and the part that takes longest to reach by hand. Recent likes are easy to manage yourself.
What are filler actions and why do they slow it down?
Scrolling the feed, watching a reel, opening DMs. They are not doing any work, and that is the point: an account that only ever performs one kind of action, back to back, does not look like a person using an app.
Do I have to log in every time?
No. The session is stored after the first run. If it stops working, delete the stored session and sign in again.
Is it safe to share my log output in an issue?
Only after you remove your username, any post identifiers and anything resembling a token or cookie. An issue is public and permanent.
Contributions are welcome. Read CONTRIBUTING.md and CODE_OF_CONDUCT.md first. By contributing you agree to license your work under the AGPL-3.0.
Found a vulnerability? Do not open a public issue. Follow the private process in SECURITY.md.
IG-Munliker is licensed under the GNU Affero General Public License v3.0 (AGPL-3.0), together with the supplemental terms in the NOTICE file. In short:
- You may use, study, modify, redistribute and even make money with this work for free, as long as you keep the complete source code available under the AGPL-3.0, including for any hosted, SaaS or network use (AGPL Section 13), and you preserve the author attribution below.
- To use this work in a closed-source or proprietary product, or to run it as a closed SaaS, you need a separate written commercial license, which may include a royalty or revenue share. See NOTICE, Section 8, and contact me.
Per AGPL-3.0 Section 7(b), the following attribution must be preserved, visibly and unmodified, in any copy, fork or deployment of this project:
Miabeyefendi (Mustafa Ihsan Albayrak) - https://github.com/Miabeyefendi
This software is provided "as is", without warranty of any kind. You run it entirely at your own risk and are solely responsible for your own use, including compliance with the terms of service of any third-party platform it interacts with, notably Instagram. Instagram is not affiliated with or endorsing this project; its name and trademarks belong to their owner. The author accepts no liability for account bans, restrictions, data loss or any other damages, to the maximum extent permitted by applicable law. Full terms are in the LICENSE and NOTICE files.
- GitHub: @Miabeyefendi
- For commercial licensing or revenue-sharing enquiries, reach me through my GitHub profile.
