The Official Power User Tips, Tricks & Productivity Thread

[Hood Critic]

@ Karakeep - where has this been

 

[Heimdall]

Yessir. You can also edit/create custom scores as well. It's a time suck tweaking things but works out in the long run

Ahhh interesting. Probably excessive for my needs, but good to get a better grasp of funnels. It sounds like a traditional reverse proxy exposure levels, but also using tailscale on the backend so likely harder to screw things up re:security

Urgency is a numeric score Taskwarrior assigns to each task. It is based on a dozen or more factors that combine to form an overall urgency score. Now this is not going to be a complete and accurate representation of urgency for many reasons, primarily because each of us has different criteria when judging the importance of a task. But it will approximate urgency, and that is the goal.

Taskwarrior uses a polynomial expression to calculate urgency. This is a simple sum of terms, where each term represents some measurable quality of a task. For example, the priority of a task is an indicator of its urgency. A high priority is more urgent than a medium priority and so on. Even a low priority is more urgent than no priority. Priority settings should therefore contribute to the urgency score.

And yeah, the tweaking -_- all these things end up taking so long to set up just right.

Yes, I saw it a simpler way of setting up a reverse proxy (it can be used to serve files too).

just causally dropping the name of an app that should have a post made about it

never heard of it before the search functions looks limited but looks good overall . whats the memory usage and disk space requirements like?

Karakeep

The Bookmark Everything app. Hoard links, notes, and images and they will get automatically tagged AI.

karakeep.app

The Bookmark Everything App​

Quickly save links, notes, and images and karakeep will automatically tag them for you using AI for faster retrieval. Built for the data hoarders out there!
Try Demo
GitHub

I already have a list of things I've been sitting on so I don't spam the thread, but it never occurred to me to include this.

It's been very useful to me. I have been trying for years to solve my tab hoarding problem (and yes, there are some I have had open for years ), and this is part of my solution. There's been... some progress

I initially tried to set up Wallabag but the version of Docker on my NAS is too old... I then tried this, then eventually got it working on my homelab.

I use it primarily as a read-it-later service, and use the lists instead of the tags.

The search seems pretty good to me - it archives the text of the hoarded link so searches go through titles and body text. I have assigned very little resources to this (like 1 core, 2GB RAM, 10GB storage) and it works fine, though CPU usage spikes when I hoard a YouTube link. The Docker images themselves take up a couple of gigs but I don't think the saved pages take up much space - atm I have just under 150 links saved, and my last backup was just under 110MB.

Spoiler

There are also some search operators you can use. (the documentation is great btw, this dev is )

I assume the computational work for the automatic tagging is done remotely since you need an OpenAI API key or Ollama URL for it, so it probably? wouldn't impact local resource use.

 

[Heimdall]

dnGrep – Powerful search for Windows

Powerful search for Windows

dngrep.github.io

dnGrep allows you to search across files with easy-to-read results. Search through text files, Word and Excel documents, PDFs, and archives using text, regular expression, XPath, and phonetic queries. dnGrep includes search-and-replace, whole-file preview, right-click search in File Explorer, and much more.

and I use this all the time. Works well even on a mapped drive.

 

[bnew]

And yeah, the tweaking -_- all these things end up taking so long to set up just right.

Yes, I saw it a simpler way of setting up a reverse proxy (it can be used to serve files too).


I already have a list of things I've been sitting on so I don't spam the thread, but it never occurred to me to include this.

It's been very useful to me. I have been trying for years to solve my tab hoarding problem (and yes, there are some I have had open for years ), and this is part of my solution. There's been... some progress

I initially tried to set up Wallabag but the version of Docker on my NAS is too old... I then tried this, then eventually got it working on my homelab.

I use it primarily as a read-it-later service, and use the lists instead of the tags.

The search seems pretty good to me - it archives the text of the hoarded link so searches go through titles and body text. I have assigned very little resources to this (like 1 core, 2GB RAM, 10GB storage) and it works fine, though CPU usage spikes when I hoard a YouTube link. The Docker images themselves take up a couple of gigs but I don't think the saved pages take up much space - atm I have just under 150 links saved, and my last backup was just under 110MB.

Spoiler

There are also some search operators you can use. (the documentation is great btw, this dev is )

I assume the computational work for the automatic tagging is done remotely since you need an OpenAI API key or Ollama URL for it, so it probably? wouldn't impact local resource use.

thanks for info.

I created a basic bookmarklet and userscript that inserts a easy UI for the search filters based on the search operators. needs dynamic values fr lists and tags though.

bookmarklet:


pastebin.com/dbv1tt2i

userscript:



pastebin.com/03AGznkC

 

[bnew]

What does this userscript do on stealthgram.com?
Stealthgram.com is a privacy-focused, web-based tool that allows anyone to anonymously view and download Instagram stories, posts, reels, and highlights without logging in or needing an Instagram account12345. Users simply enter a public Instagram username, and the platform fetches the content, allowing for easy browsing, high-resolution downloads, and a user-friendly, mobile-compatible interface.

---

This userscript ("Stealthgram Carousel Media Gallery Toggle") adds extra functionality and convenience to how you interact with carousels (multi-image/video posts) on stealthgram.com.

---

Simple Overview​

• Adds a floating toggle button to the website.
• When turned ON, it scans any visible Instagram-style carousel and extracts all images and videos, no matter how many slides there are.
• Shows all collected media in a special "gallery" modal overlay—separate from the Stealthgram page content.
• In the gallery, you can:
  • See all carousel images and videos as easy-to-click thumbnails.
  • Click on a thumbnail to preview a larger version without leaving or reloading the page.
  • Click the preview image/video to toggle between regular and high-quality versions (if available).
• Closes and cleans up everything (button, gallery, logs) when you toggle OFF or navigate away.

---

Step-by-Step, In Simple Terms​

1. Button appears:
   A floating button appears at the bottom right of Stealthgram pages.
2. Extraction on demand:
   When you click this button, the script:
   • Looks for the active Instagram carousel (a multi-image or video post).
   • Clicks through every slide, one-by-one, collecting the URL for every unique image or video it sees.
   • Stops early if no new content is found for three cycles, so it never hangs or loops forever.
3. Gallery display:
   When done, a pop-up overlay ("gallery") appears, showing all collected images and videos as thumbnails.
4. Preview features:
   • Click a thumbnail to see it larger in the gallery viewer.
   • Click the large preview image to instantly switch between normal and maximum quality (if available) without leaving the gallery.
   • Videos play right in the gallery.
5. Easy cleanup:
   Turn the button OFF at any time, and the gallery and button instantly disappear, leaving Stealthgram as you found it.
6. Console log:
   At every gallery session, the script logs helpful troubleshooting info, showing exactly which items were collected or previewed.

---

What Makes This Userscript Unique?​

• Automated Carousel Extraction:
  Instead of manually clicking through images or videos, the userscript does this for you—faster and more reliably.
• Smart Early Exit:
  It notices if it's seeing the same media over and over, and will stop collecting after three cycles with no new images or videos, saving you time.
• High-Quality Downloads:
  When possible, you can preview and save higher-resolution versions of images directly from the gallery overlay.
• Non-intrusive:
  The script doesn't change or break Stealthgram's existing features. All enhancements are accessed only through the toggle button and leave the page untouched when turned off.
• Troubleshooting Information:
  All activity is logged clearly in the browser console for transparency and easy debugging if something doesn't work as expected.

---

Summary Table​

Feature	What it Does
Floating Toggle Button	Turns the gallery feature ON or OFF
Carousel Auto Extraction	Collects all carousel images/videos automatically
Early Stop on Duplicates	Stops after 3 cycles with no new content, avoiding infinite loops
Modal Gallery Overlay	Shows all collected media as easy-to-browse thumbnails/previews
Preview Quality Toggle	Click preview to swap between normal and full-res (if available)
Console Logging	Logs key actions for user troubleshooting and transparency
Clean On Deactivate	Cleans up all UI and listeners when turned OFF or on page exit

---

In short:
This userscript gives you a powerful, one-click way to instantly collect, preview, and access all images and videos in Stealthgram carousels—saving you clicks, time, and providing a seamless, private gallery experience right on the site.





pastebin.com/n6qEiZGn

 

[bnew]

i'm always archiving links but i wanted another way instead of inserting save links to pages.
initially written by grok3 and updated by ernie 4.5 by baidu
check the browser console for the saved links.

LinkArchiver is a bookmarklet—a small JavaScript tool you save and run in your browser—that helps you quickly collect, filter, and archive web links from any web page or from a list you paste yourself, sending them to the Internet Archive's Wayback Machine for long-term preservation.

---

What does LinkArchiver do?​

• Collects Links: It can automatically gather all links to webpages, images, and videos found on the currently open page, or work with a list of links you paste in.
• Filters Links: You can use a regular expression (regex) pattern to include only the links you care about (e.g., links to a certain domain or with a specific path).
• Transforms Links: Add rules to modify domains or paths before archiving (for example, swap "olddomain.com" for "newdomain.com" in all collected URLs).
• Shows Suggestions: The tool suggests patterns and example links to help you quickly set the right filter.
• Bulk Archives: Once your links are selected and prepared, LinkArchiver submits each link to the Wayback Machine for archival.
• Handles Rate Limits: Warns you if you try to archive more than 100 links (the daily limit for anonymous Wayback submissions).
• Pause and Resume: While archiving, you can pause and resume the process.
• User Feedback: It shows notifications ("toasts") for progress, errors, and completion results.

---

User Interface (UI) – What does it look like?​

When you run LinkArchiver, a pop-up panel appears in the middle of your browser window with these main parts:

• Title: "LinkArchiver"
• Mode Selection:
  • "Scrape from current page" (auto-collect links from the open page)
  • "Use pasted links" (paste your own list into a textbox)
• Paste Area: (only if you select "Use pasted links")
  • Text box where you can paste a list of URLs (one per line)
• Regex Pattern Field:
  • Optional. Enter a pattern to filter the links you want (e.g. .*example\.com.*)
• Replace Rules:
  • Section to add one or more "replace" instructions to change the domain or path of collected links before archiving
• Suggestions:
  • List of suggested example links and regex patterns, clickable for easy selection
• Controls/Buttons:
  • "Start Archiving": begin the archival process
  • "Pause"/"Resume": pause/resume bulk archiving
  • "Close": exit the tool
  • "+ Add Rule": add more replacement rules
• Warnings:
  • Notification if you exceed the Wayback Machine’s daily anonymous limit

Example UI (visualized as text, based on code):​

text
+------------------------------------------------------+
| LinkArchiver |
| [x] Scrape from current page [ ] Use pasted links |
| Regex Pattern (optional): [ ________ ] |
| Replace Rules (optional): |
| [Domain | Path] [what] [with] [+ Add Rule] |
| [Suggestions / Patterns appear here] |
| (Matching Full URLs, Suggested Regex) |
| [Start Archiving] [Pause] [Close] |
| Rate limit warning (if >100 links) |
+------------------------------------------------------+

Example: How a user might interact​

1. Collect links from the current page:
   • Select "Scrape from current page"
   • The suggestions box quickly lists sample links and suggested regex patterns matched to the page.
2. Filter links:
   • Enter a regex like ^https?://(www\.)?example\.com/.* to only keep links from example.com.
3. Modify links:
   • Add a rule:
     • Type: "Domain"
     • Replace "olddomain.com" with "newdomain.com"
4. Start archiving:
   • Click "Start Archiving".
   • Toast appears: "Archiving 1/35: https://example.com/page1"
   • Progresses through each link, obeying rate limits and allowing pause/resume.

---

How does LinkArchiver work in simple terms?​

• It lets you easily select and prepare a batch of links for archival—either from the website you’re viewing or from a list you provide.
• You can filter (choose which links to keep), edit (fix or swap domains/paths), preview, and then send all those links to the Wayback Machine for permanent saving.
• It guides you with friendly suggestions and notifies you of limits, errors, and completion—all inside a simple panel right on the webpage.

---

In summary:
LinkArchiver is a convenient power tool for anyone who wants to mass-archive web content, making sure important pages and media don’t disappear from the internet. It handles scraping, filtering, bulk actions, link editing, and user feedback through an easy-to-use pop-up, right in your browser, no special install required.



pastebin.com/zDzvgpKe

less comprehensive scraping:




pastebin.com/SWznQu8L

 

[Macallik86]

Being that we all have access to LLMs, perhaps it might be more engaging to have a bit more of your perspective and a bit less of your LLM's output in the thread btw bnew?

 

[bnew]

Being that we all have access to LLMs, perhaps it might be more engaging to have a bit more of your perspective and a bit less of your LLM's output in the thread btw bnew?

i'm interested in people sharing their custom solutions for their workflows. not a lot of people are sharing their output. some of the stuff I post here I use on thecoli regularly or making thecoli more useful was a motivator. I hate seeing dead links usually images, tweets and other embeds so I get the LLM to write code that allows me to mirror/backup data in a myriad of easy ways.

 

[bnew]

This userscript automatically adds an extra link ("Stealthgram link") directly below Instagram posts that are embedded in forum threads on XenForo-based sites (like TheColi.com and others). The added link allows users to open the same Instagram media (photo, video, reel, etc.) anonymously and directly on Stealthgram.com1. Stealthgram is a tool for anonymously viewing and downloading public Instagram content.
Detailed Summary:

• What Does It Do?
  • When you view a forum thread with Instagram media embedded (using the s9e embed style, common in XenForo forums), this script scans for those Instagram posts.
  • Beneath each Instagram embed, it adds a new text link labeled "Stealthgram link."
  • Clicking this link opens the same Instagram post on Stealthgram.com – a site where you can view and (sometimes) download Instagram posts anonymously, with no login needed1.
  • The script works automatically, even if new Instagram embeds are loaded while you scroll or while the page updates, ensuring the new links always appear.
  • It avoids adding duplicate links under the same embed.
• Why Use This?
  • Anonymous viewing: You can view Instagram media without logging in or letting Instagram know who you are.
  • Site reliability: If Instagram embeds are broken or blocked, Stealthgram often provides an alternative way to see the media.
  • Convenience: No need to manually copy-paste Instagram links into Stealthgram—just click.
• How Does It Work?
  • Detects Instagram embeds by looking for special "iframe" code in forum posts.
  • Extracts a unique media ID from the embed.
  • Inserts a link using this ID, pointing straight to the Stealthgram view for that post.
  • Monitors the page for new Instagram embeds and handles them automatically.

How to Install on Desktop Browsers (Firefox/Chrome):

1. Add a Userscript Manager Extension:
   • For Chrome or Chromium browsers: Install Tampermonkey from the Chrome Web Store.
   • For Firefox: Install Tampermonkey OR Violentmonkey from Firefox Add-ons.
2. Install the Userscript:
   • Copy the script code provided above.
   • Click the userscript manager’s extension icon and select “Create a new script” or "Add a new script".
   • Delete any default code and paste in the entire userscript.
   • Click “Save.” The userscript manager will now run this script automatically on supported sites.
3. Browse Supported Forums:
   • Go to any supported XenForo-based forum thread (e.g., thecoli.com or forums with URLs containing /threads/).
   • Instagram embeds in posts will now have a "Stealthgram link" below them. Click to view media on Stealthgram.

How to Install on Mobile Firefox (Android):

1. Add a Userscript Manager Extension:
   • Open Firefox on your phone.
   • Go to addons.mozilla.org and search for “Tampermonkey.” Install it.
2. Install the Script:
   • Tap the Tampermonkey icon, choose to create a new script, and paste in the code above.
   • Save the script.
3. Use:
   • Visit supported forums in Firefox for Android.
   • Stealthgram links will appear below Instagram embeds as they do on desktop.

Note:

• Chromebooks and some Android browsers (Chrome for Android) may not support userscripts directly.
• iOS browsers generally do not support userscripts unless you use third-party paid apps with userscript support.

No technical skills are needed beyond copy-pasting and saving within Tampermonkey/Violentmonkey.
For more on Stealthgram features and privacy, visit

'stealthGram – Instagram Story Viewer and Downloader'

Anonymous Instagram Story Viewer. View and Download IG stories, reels, posts, highlights and full size dp for any public profile!

stealthgram.com

pastebin.com/SfpwFJce
the link works but coli embed is old & broken.

 

[bnew]

Here is a full userscript that will insert the profile link, add the clipboard button, toast notifications, and a filename button under the profile name on Stealthgram media pages exactly as you described. Install this in a userscript manager like Tampermonkey or Greasemonkey.



pastebin.com/BDqv1zzZ

How it works:

• Runs automatically on any https://stealthgram.com/media/* page.
• Inserts a line under the profile name (netflix in your example) showing:
  • Profile link (clickable)
  • (clipboard button to copy profile link; works with Tampermonkey/GM or modern native clipboard)
  • "Filename" button that copies a filename string as described.
  • Toast notifications for feedback.

You can adjust selectors and styles as needed for any small DOM/layout changes. The script is robust against most dynamic page loads.

 

[bnew]

updated to support twitter embeds which won't load since it may require an account.


I incremented the version number and integrated Twitter handling cleanly into the same observer/throttle system.



pastebin.com/sEd3WRLn


What changed:

• Bumped version → 2025.08.16.
• Preserved Instagram logic exactly as-is.
• Added Twitter embed support:
  • Extracts tweet ID from iframe[data-s9e-mediaembed="twitter"].
  • Generates all requested link variants (twstalker, archive, xstalk, nitter, xcancel).
  • Inserts them neatly after the iframe with duplicate prevention (data-twstalker-done).
• Kept the same observer/throttle structure so it works dynamically on newly loaded embeds.

Do you want me to also merge both Instagram and Twitter link blocks into a single <div> container per embed, so they stay visually consistent, or keep them separate as they are now?

 

[bnew]

an update to the "Stealthgram Profile Link & Clipboard Tools" userscript

Got it The reason your profile link and copy buttons sometimes disappear is because Stealthgram loads content dynamically and may re-render the <h5.header> node after your script already inserted the buttons. That means your tryInsertTools loop sometimes finishes before the final DOM is stable, so the injected container gets wiped out.


To fix this reliably, you need to watch for DOM changes with a MutationObserver in addition to your initial timed retries. That way, even if the profile header gets replaced or reloaded after your code runs, your tools will be reinserted automatically.


Here’s your userscript updated to v1.1 — I preserved the structure and functionality, but added a MutationObserver to track dynamic changes while still keeping your retry mechanism:



pastebin.com/5SAdHzCw
the link isn't broken

Fix summary:

• Sometimes Stealthgram replaces the profile header after your tools are injected, wiping them out.
• I added a MutationObserver so the script re-runs insertTools() whenever DOM changes.
• Kept your retry loop for initial load.

This guarantees your profile link and copy buttons will persist across dynamic page updates.

 

[bnew]

whenever i play videos from an instagram embed it only plays once and META forces you to open a link to instagram to watch the video again. so I had an idea to add a reload iframe button next to the iframe so i don't have to leave the current webpage. gpt-5 figured it out in two urns basically and ijust tweaked the options further. deepseek one shotted and I just tweaked it once. gpt-5 code has options in code, deepseek doesn't. the ui is clean.

• Finds Instagram iframes (iframe[data-s9e-mediaembed="instagram"]).
• Adds a reload button only once per iframe (no duplicates).
• Places the button on the right, vertically centered.
• Clicking the button reloads only that iframe, not the whole page.
• Works for dynamically added iframes via MutationObserver.
• Runs efficiently with throttling.

To use:

1. Copy the entire code above.
2. Create a new bookmark.
3. Paste it into the bookmark’s URL field.
4. Click the bookmark on a page with Instagram iframes — reload buttons will appear.
   

---

Userscript Version (no auto-reload)​

chatgpt gpt-5


pastebin.com/cCvH3FiX


deepseek-3 0528



pastebin.com/iEKc05Mi

---