DEV Community: GBTI

Empirical Responsive Images: Auto Image Sizes, WebP and AVIF for WordPress

GBTI — Thu, 05 Nov 2026 14:49:51 +0000

By Jesus Iniesta, GBTI Network Supporter. Originally published on gbti.network.

Empirical Responsive Images finds the images a WordPress site sends too large for the space they fill, then creates the sizes that actually fit. Activate it and that's it: there is nothing to configure, no bulk regeneration to run, no queue to approve, and no account or API key.

The problem it solves

WordPress can only offer the thumbnail sizes it already has. When none of them is close to the slot an image is shown in, phones end up downloading images made for desktops. PageSpeed Insights flags this as "Improve image delivery" or "Properly size images", and it can hold back Largest Contentful Paint (LCP).

Other tools ask you to guess which sizes you need. This one measures them on your real pages and only creates the sizes your layouts use.

How it works

A small browser check compares each displayed image with the sizes already available. Images with a good match are ignored.
A missing size has to be confirmed before any work is scheduled. By default, regular visitors confirm it across three separate 15-minute periods.
WordPress creates the new files later as a background task, up to 50 image jobs a day.
The normal responsive image markup (srcset and sizes) picks them up, and the page cache is refreshed.

Nothing is resized in the visitor's browser, and no image is generated while a visitor waits for the page.

What you get

Missing thumbnail sizes created automatically, without regenerating the whole media library
WebP and AVIF versions when the server's image library supports them, kept alongside the originals
Coverage for Media Library images and local theme and plugin images
Cache refresh for WP Rocket, W3 Total Cache, WP Super Cache, LiteSpeed Cache, Breeze, Autoptimize and SiteGround Speed Optimizer
Works with lazy loading, page builders and WooCommerce galleries that output normal WordPress image elements
Optional status tables under Tools > Responsive Images, plus WP-CLI commands to list learned sizes or force a rebuild

Custom crops, image choices and picture layouts stay as they are. The plugin fills proven gaps rather than replacing WordPress responsive images.

Privacy and limits

Everything stays on your site: no images are sent to a CDN or an outside image service. Reports carry only a code for a local image and a rounded missing size, without the page URL, visitor identity, IP address or browser details. Visitors can't submit arbitrary image URLs or invent sizes, and hard limits on requests, queued jobs, retries, daily work and free disk space keep generation bounded.

Result

On one test page, the mobile PageSpeed score went from 75 to 99 and LCP improved from 7.2 s to 2.3 s. Results depend on the theme, original images, cache and hosting, so no score is guaranteed.

Free and open source under GPLv2 or later. Requires WordPress 6.9+ and PHP 7.4+.

The resource shared in this post is a new project published by GBTI Network member Jesus Iniesta.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow Jesus Iniesta's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/gsusi/

A Crystal Ball for the Next Digital Agency: How agentic project management will change the agency-client relationship and revenue model.

GBTI — Thu, 29 Oct 2026 14:09:46 +0000

By @atwellpub, GBTI Network Supporter. Originally published on gbti.network.

Listen to the podcast version (22 minutes and generated by Notebook LLM. The podcast version only skims high level concepts, while the paper breaks down a product projection map in detail)

A Strategic Vision for the Next Generation of Digital Agencies; First published on Google Docs and then shared on LinkedIn.

Executive Summary

This paper presents a vision for the next generation of digital agencies, where AI-enabled client operating environments transform how web properties are managed, how technical capabilities are delivered, and how agencies build sustainable, recurring service relationships.

In this paper, we propose a future in which agencies manage client projects through an integrated project management environment, combining project coordination, resource management, and client relationships with conversational AI interfaces. Through an agency-managed MCP architecture, clients can interact directly with their websites and connected systems, using capabilities and permissions configured and governed by their agency.

Traditional CMS administration interfaces remain accessible alongside these capabilities, allowing clients and developers to perform specialized tasks, make visual edits, and interact directly with the underlying systems when appropriate.

The agency's responsibility evolves from performing individual development tasks to configuring, securing, maintaining, and expanding the client's AI-enabled infrastructure.

The following sections explore how this vision will most likely take shape, beginning with the proposed client operating environment and the integration of frontier AI models and MCP capabilities. Particular attention is given to the permissions and governance frameworks through which agencies manage access, delegate authority, and control the capabilities available to clients, their teams, and AI agents.

We then examine the evolving role of agencies as providers of managed digital infrastructure, including the opportunities for recurring revenue and continued professional service provision.

Finally, we consider the technical challenges of agent-driven operations, the architectural principles necessary to support these systems, and the broader strategic implications for the agency industry.

Together, these considerations establish a proposed direction for the next generation of agency-driven web property management, while raising important questions about who will develop, deliver, and ultimately lead this emerging market.

Who will build? Who will go to market? Who will win?

To read the article version of the paper visit: https://gbti.network/articles/crystal-ball-for-the-next-digital-agency/

And to read the original published Google Doc complete with SWOT analysis visit:
https://docs.google.com/document/d/1UeDXC09Q68NtVx0vdGMvvznqC0CdVwStX1g3lUprY50/edit?usp=sharing

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow @atwellpub's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/atwellpub/

Slashy: AI Email Client with an MCP Server and Agent Skills

GBTI — Tue, 27 Oct 2026 01:33:03 +0000

By @GaddipatiHarsha, GBTI Network Supporter. Originally published on gbti.network.

Your inbox, finally on autopilot

Most "AI email" tools bolt a chatbot onto Gmail. Slashy is the whole email client, rebuilt around an assistant that actually does the work. Y Combinator's directory sums it up in three words: Cursor for Email.

Open a thread and the reply is already waiting, written the way you write. Slashy reads the full conversation, drafts the response you would have sent, chases the follow-ups you would have forgotten, and books the meeting without the five-email back and forth. You review, tweak if you want, and hit send.

What users say

Slashy is rated 4.9/5 on G2. A few lines from verified reviews:

"Slashy is the best agent-native email client I've used, and I've probably given almost all of them a serious try."
Adam B., Founder / CEO

"The MCP connector, so badly missing with many other email clients, allows me agentic access to my entire mailbox."
Verified user in Investment Banking

"It picks up on my voice and what I like and turns out replies I send with barely any edits... Delegating that stuff saves me hours every week."
Julian J., Technical Support Representative

"I actually started a side project to just create my own program, then I discovered Slashy. It was literally everything I wanted in an email client."
William R., President

What it does for you every day

Drafts in your voice. Not generic AI filler. Slashy learns how you write and drafts replies that sound like you on a good day.
Inline rewrites. Highlight any text to rewrite it, shorten it, or change the tone.
Scheduling on autopilot. It checks your calendar, proposes times, and creates the event.
Follow-ups that don't slip. Reminders and automations nudge threads forward before they go cold.
Meeting prep and contact research. Walk into every call knowing who you're talking to and what you last discussed.
Everywhere you work. Mac, iPhone and web, plus an iMessage agent and Slack access. Gmail and Outlook, with unlimited inboxes.

Built for people who live in Claude Code, Cursor and Codex

This is the part GBTI readers will love. Slashy runs a hosted MCP server at https://slashy.ctrlcenter.ai/mcp, so your coding agent can work your inbox directly: draft, send, reply, schedule, label, check availability, research a contact, prep a meeting and set reminders. It connects over OAuth 2.1 with PKCE, so there are no API keys to copy, rotate, or leak.

Want more than raw tools? Two open repos ship the agent side:

Slashy-com/skills: seven agent skills in the open SKILL.md format, covering onboarding, inbox briefings, a reply and follow-up sweep, meeting-to-email, relationship briefs and automation setup. Install with npx skills add Slashy-com/skills, with published install paths for Claude Code, Cursor, Codex, Copilot in VS Code and Windsurf.
Slashy-com/slashy-plugin: a Claude plugin for Gmail or Outlook mail and Google or Outlook calendar.

Try asking your agent "triage my inbox and draft replies to anything urgent" and watch it go. The skills are built to read full context before writing, default to drafts when sending intent is unclear, and treat email content as untrusted input rather than instructions.

Pricing, plus 20% off for GBTI readers

Professional is $25 per user per month billed annually, or $30 monthly. That covers intelligent drafts and inline rewrites, calendar and scheduling, automations and reminders, team collaboration and unlimited inboxes. Enterprise adds SSO and SCIM, a dedicated account manager, guided onboarding, and custom security and data controls.

GBTI readers get 20% off: sign up through slashy.com/?ref=GBTI or enter the code GBTI at checkout. Every plan starts with 7 days of full access.

Enterprise-grade security

Your email is the most sensitive data you have, and Slashy treats it that way. SOC 2 Type II and CASA Tier 2 assessments are complete. Customer data is never used to train Slashy's models or any third-party models, and AI providers are contractually held to the same. Data is encrypted in transit with TLS 1.2+ and at rest with 256-bit AES, hosted on AWS in the United States. Disconnect and your production data is gone within 24 hours.

Who makes it

Slashy is a Y Combinator Summer 2025 company based in San Francisco, founded by Harsha Gaddipati (CEO) and Dhruv Roongta (CTO). Got a feature request? Leave a comment below, the founders read them.

The resource shared in this post is a new project published by GBTI Network member hgaddipati1118.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow hgaddipati1118's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/hgaddipati1118/

NameDaemon: Find .com Names From the Browser or the Command Line

GBTI — Mon, 26 Oct 2026 13:26:43 +0000

By Krasen Slavov, GBTI Network Supporter. Originally published on gbti.network.

NameDaemon is about thirty free tools for working with .com domain names, and all of them read from one daily snapshot of every registered .com. Its own home page puts it as "30 tools, one snapshot." The site is available in English, Spanish, French and German.

A daily snapshot of every registered .com

Each morning the full .com zone file is downloaded from ICANN's Centralized Zone Data Service, published by the registry operator, Verisign. It holds names only, with no owner and no contact details. Today's list is compared against yesterday's: a name in yesterday's and not in today's has left the zone, and a name in today's and not yesterday's has appeared. That is how a drop gets spotted. For the names that matter the live registry status is then checked over RDAP, because a name missing from the zone is not always free.

Every result says which day it is from. The list of names itself is never published and there is no bulk export.

The daily numbers are published on their own page. On 2026-10-20 the zone held 167,843,117 names, 67,718 more than the day before, with 189,643 names appearing and 121,925 leaving. The median name was 12 characters, 9.93% contained a hyphen, and GoDaddy ran 21.58% of the zone by first nameserver.

Finding, judging and watching a name

The tools come in three groups.

Finding a name: a generator, short pronounceable names that are missing from the zone, domain hacks, a handle matcher that checks GitHub, npm, PyPI and Bluesky, ENS and Solana names read from the blockchain, and a brand studio.

Judging a name: a 0 to 100 score that shows every part of its reasoning, an appraisal that reports strength and deliberately gives no price, a one-page Domain Report you can share, a screener and cost lookups.

Watching a name: an audit of what a name used to be, read from the Internet Archive, plus lookalikes, a health check covering expiry, transfer lock, DNS, SPF, DMARC and blocklists, a WHOIS lookup, a drop calendar, deleted domains, keyword trends and hosting stats.

Anyone can watch one name free and get an email when it frees up. Saved domains stay in your own browser.

Packages, the command line and the API

The same tools are packages, free under MIT and needing no key.

npm install namedaemon
pip install namedaemon

The npm and PyPI packages expose the same typed methods with no dependencies at all: Node 18 and later, Deno and Bun on one, Python 3.9 and later on the other, using nothing but the standard library. Clients for Go (with a Homebrew tap), PHP (Composer) and Ruby (RubyGems) sit beside them. There is a command line too, through npx namedaemon or the namedaemon binary, with --json for the raw answer.

npx namedaemon rate example
npx namedaemon ideas nova 10
npx namedaemon check nova novax mybrand
npx namedaemon tco io

The packages wrap a public JSON API, so any language can call it with a GET or a POST:

curl "https://namedaemon.com/wp-json/nd/v1/rate?name=example"

The site runs on WordPress, with a theme of its own, and the API is served straight out of it. A status panel on the developer page reports the API up, checked every five minutes, at 99.97% over 30 days and 99.85% over 7, measured from outside by UptimeRobot.

The extension, the desktop app and the web app

A browser extension checks any .com from the toolbar. Type a name, or select one on any page and right-click it, and it says whether the name is taken, free or about to drop, with the estimated release date and where to register or backorder it. From there you can appraise it, check its health and audit its past without leaving the tab. It is free, needs no account, and runs on Chrome, Edge, Firefox, Brave and Opera.

A desktop app for Windows and Linux puts nine of the tools in a window of their own: the Domain Name Generator, Availability Checker (up to 50 names at once, with CSV export), Domain Report, Domain Score, Domain Appraisal, Domain Audit, Domain Health Check, Short Domains and the Drop Calendar. It adds a watchlist kept on your computer that sends a notification when a name changes status, and a tray icon for checking the name you just copied. It is free and open source under MIT, has no account, no ads and no analytics, and sends a name to NameDaemon only when you check it. It is not code-signed yet, so Windows may show a warning before it installs. Download it from SourceForge or the Snap Store, or directly from namedaemon.com, where every file has a SHA-256 checksum.

The same tools also work as an installable web app, from the browser or from the Microsoft Store. It runs under Chrome and Edge on Windows, macOS, Linux and Android, and on Safari through Add to Home Screen.

What it costs, and what Pro will be

Every tool, package and app is free, with no account and no key.

Pro is not open. Its page is headed "PRO / COMING SOON" and carries a waitlist where a checkout would be. The planned price is from $12 a month, and the page says in as many words that the price is not final until Pro opens.

What it is planned to carry: up to 15 keyword rules matched against each day's new and departing names, one daily digest by email, Telegram or on the site with the last 30 days searchable, watching up to 50 single names, saved domains kept on the account and not only in the browser, keyword trend alerts on up to 10 words that report counts and never a list of names, following the twelve weekly domain lists, 30 WHOIS lookups a day without the verification step visitors go through, and AI notes on each match. Payment through Stripe, with a sign-in link by email instead of an account and a password.

Where the numbers stop being reliable

A Limits page says where, in plain words and not in small print. A name missing from the snapshot is not necessarily available: it can be registered without nameservers, or registered after the snapshot, so only the live registry check says a name is free and you should still confirm with your registrar. Drop dates are estimates worked out from the registry's last-changed date and can be off by several days, and popular names are taken within seconds of dropping by automated backorder services.

Scores are heuristics rather than valuations, which is why the appraisal reports strength and gives no price: a price needs sales data the site does not have. The audit reads what the Internet Archive happens to hold and the health check reads public registry and DNS data, so a clean result means nothing was found, which is not the same as nothing having happened, and neither is a security audit. The offensiveness screener covers 20 languages written in Latin letters and misses slang. Alerts and watches run once a day, so a watch is a head start and not a guarantee.

The page ends on the trademark question. A name being available does not mean you may use it.

The resource shared in this post is a new project published by GBTI Network member Krasen Slavov.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow Krasen Slavov's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/krasenslavov/

MadTicket Brings Event Discovery and Ticketing Together in Spain

GBTI — Fri, 23 Oct 2026 23:36:35 +0000

By Michel Moraes, GBTI Network Supporter. Originally published on gbti.network.

MadTicket sells tickets to concerts, parties, festivals and shows across Spain, and gives the organisers who run those events the tools to sell them.

Buying a ticket

Buyers browse by city, Madrid, Barcelona, Valencia, Sevilla and Bilbao, or by category, and search by artist, event or venue. Checkout takes a card or Apple Pay, and any organiser discount applies on its own. Tickets carry a QR code, live in "Mis entradas" and can be added to Apple Wallet. When an event is sold on the organiser's own site, MadTicket says so and links there instead. There is an iPhone app; the Android version is listed as coming soon.

Publishing an event

An organiser picks one of two routes.

A free listing puts the event in the MadTicket agenda with a button through to the organiser's own sales page. It costs nothing and carries no commission. It buys no advertising and promises no number of visits.

Selling through MadTicket runs the checkout: Stripe for payment, a unique QR code on every ticket, and billing details from the organiser before it can be switched on.
Either way the event is reviewed before it appears. The review checks the city, the date, the purchase link, the organiser, whether it duplicates an existing listing, and whether the texts and images are the organiser's to use. An organiser who finds their own event already listed by somebody else can ask to take it over.

Prices, coupons and affiliates

Ticket types with their own price, capacity and visibility: General, VIP, early bird, whatever the event needs.
Coupons with a percentage off, a usage limit and an expiry date, optionally restricted to certain ticket types.
Combo discounts, such as two VIP plus one General.
An affiliate programme: links with a fixed or percentage commission, limits and expiry per link, and tracking of how many tickets each one sold.
Second dates for an event, with courtesy tickets for people who came to the first.
Refunds through Stripe, which warn first if the ticket was already scanned or the event has ended.

Running the event in Backoffice

MadTicket Backoffice is a native app with a responsive web version. It shows revenue, commission and settlement together, and tickets sold against capacity as they go. Orders can be searched, filtered and worked through from a phone on the night. The whole order list exports to PDF or CSV, grouped or filtered by ticket type, with the buyer name, email, reference, amount and date.

Scanning at the door

MadTicket Scan is native on iOS, with a web scanner for Android, and a QR can be scanned from any mobile browser. Validation is instant and shows the buyer's name. A ticket presented twice is blocked, and the screen gives the date and time it was first scanned. Nothing is printed and no extra hardware is needed.

What it costs an organiser

There is no monthly fee and publishing costs nothing. A commission applies per ticket sold, set by the organiser's plan, and the rest reaches them through Stripe. Settlement is requested once the event has finished. The rates themselves are not published on the site.

Search across two languages

A catalogue that mixes Spanish and Portuguese breaks keyword search: a search for "torneo deportivo" returns nothing for an event called "Torneio de Pádel". MadTicket closes that gap with embeddings. Events are embedded and stored in Postgres with pgvector, keyword search still runs first, and a search by meaning using cosine distance runs only when keywords return fewer than three events. Event pages show similar events from the vectors already stored, in a single query. A LinkedIn post on 16 October 2026 covers how it was built and measured, including the four models compared and why a fixed similarity threshold did not work.

Getting in touch

Thank you for reading this far. MadTicket is listed on the GBTI Network, a co-op where developers publish what they build and talk about how it was built. The author profile, linked from the byline at the top of this page, carries the rest of the work and is the place to take this further.

The resource shared in this post is a new project published by GBTI Network member Michel Moraes.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow Michel Moraes's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/michelmany/

IsHacked combines external malware detection, uptime monitoring, SEO checks, SSL alerts, and email-health testing

GBTI — Thu, 22 Oct 2026 00:36:33 +0000

By Jesus Iniesta, GBTI Network Supporter. Originally published on gbti.network.

IsHacked checks websites from the outside, the way visitors and Google see them. It catches the hacks owners miss because the site looks normal to them: spam served only to Googlebot, redirects that fire only for phones arriving from Google, malware loaders hidden from logged-in admins. It also watches the everyday failures that hurt just as much: downtime, PHP errors, expiring SSL and broken email. There is nothing to install and it needs no access to the site. The free check takes about a minute, with no sign-up.

What it checks

25 checks per site, from Germany, the US and the UK:

Hacks and malware: redirects that only phones, Google visitors or one country see; malware signatures such as NDSW, Balada and SocGholish; new script, iframe and form domains; rogue WordPress users and fake plugins; exposed database dumps, wp-config backups, .env and .git.
SEO spam and indexing: cloaking (what Googlebot sees versus visitors), casino, pharma and Japanese keyword spam, hidden links, injected sitemaps, robots.txt tampering and surprise noindex. Inner pages sampled from the sitemap get the same checks.
Uptime and errors: pings every 5 minutes with down and back-up alerts confirmed from a second location, 5xx and PHP fatal errors, WordPress critical-error screens, database errors, stuck maintenance mode and suspended-account pages.
SSL, domain and DNS: certificate and domain expiry, nameserver and MX changes, and the http to https redirect.
Email deliverability: SPF, DKIM and DMARC matched to the mailbox provider, plus Spamhaus blocklists.
Security hygiene: outdated WordPress core and end-of-life PHP, mixed content, version leaks, username enumeration, XML-RPC and security headers.

How it works

Run a free check. Enter any URL at ishacked.com. The report is ready in about a minute and gets a shareable link for 7 days.
Add your sites. Paste client URLs one per line, with an optional label. The first scan learns what is normal for each site, so after that only changes are reported.
Hear only when something changes. Plain-English alerts go to you and up to 5 extra addresses, or to Slack, Discord or a webhook, followed by an all-clear when it is fixed.

For developers and agencies

Raw headers, the HTML as Googlebot and as a phone, redirect chains and the full JSON of every scan. A REST API to trigger checks from CI after a deploy, CSV export, and a one-click re-check of every site.

What it can't do

No external scanner can prove a site is clean: a backdoor that shows nothing to the public stays invisible from outside. IsHacked catches what visitors and Google see, so pair it with a server-side scan of files and the database.

Pricing

Priced near cost. Every plan gets every check, and you can cancel any time:

Solo, $3/month: up to 10 sites
Pro, $9/month: up to 50 sites
Agency, $25/month: up to 200 sites

The resource shared in this post is a new project published by GBTI Network member Jesus Iniesta.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow Jesus Iniesta's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/gsusi/

Turn Spoken Family Recipes Into a Cookbook: Culinary Lineage

GBTI — Mon, 19 Oct 2026 13:28:40 +0000

By Jesus Iniesta, GBTI Network Supporter. Originally published on gbti.network.

Culinary Lineage turns family recipes told out loud into a cookbook. A cook talks through a dish the way they remember it, and Culinary Lineage puts the ingredients, steps and stories into a clear recipe, asks about anything unclear, and adds nothing that was not said. The first recipe is free, with no card or account.

How it works

Start a book or invite family. Record your own recipes, or send Mum, Grandma, Dad or Grandad a private invitation, now or on a special day.
Tell it naturally, in any order. The invitation opens in the browser on a phone, tablet or computer. There is one large record button, or they can upload a voice note. No app to install and no password to remember.
Only answer what is missing. Ingredients, steps and memories are sorted into place. If "the little blue cup" needs a measurement or the oven timing does not add up, Culinary Lineage asks a short question instead of guessing.
Check each recipe and build the book. Add family photographs, arrange the chapters and export a high-resolution PDF.

Made to print

The PDF is US Letter with safe margins, so it can be printed at home, at a local print shop or with an online book printer such as Lulu or Blurb, in as many copies and bindings as the family wants. An optional check by the team covers layout, page breaks and image quality before printing. The physical book is not included in the price.

Private by default

There are no public profiles and no recipe marketplace. The family owns its recordings and recipes and can export or delete them from the book settings. If the connection drops mid-recording, the unfinished recording stays on the device.

Pricing

Each book is a one-off payment, with no subscription and no charge for extra recording time:

The Keepsake, €69: up to 20 recipes from 1 family member
The Family Book, €109: up to 50 recipes from up to 3 family members
The Legacy Edition, €169: up to 100 recipes from up to 6 family members

If nobody has recorded a recipe yet, a full refund is available within 30 days of purchase. A book can also be given as a gift.

Languages and guides

The site is available in English, Spanish and Russian. Free guides cover preserving family recipes before they disappear, questions to ask parents and grandparents about their cooking, and planning a family cookbook from first recipe to printed book.

The resource shared in this post is a new project published by GBTI Network member Jesus Iniesta.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow Jesus Iniesta's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/gsusi/

Jockey Radio Streams Internet Radio Through Minecraft Jukeboxes

GBTI — Sat, 17 Oct 2026 15:32:32 +0000

By @gbtilabs, GBTI Network Supporter. Originally published on gbti.network.

Jockey Radio is a new Minecraft Fabric server mod from GBTI Network that turns ordinary music discs into live audio sources. Drop one into a jukebox and nearby players can hear internet radio, HLS streams, local audio, or an incoming RTMP broadcast through Simple Voice Chat.¹

Server operators manage everything through /jockey, creating discs, setting their range and appearance, monitoring stream status, and even running several stations at once. OBS, SavePoint, and other broadcast tools can push directly into the Minecraft server, while SavePoint stations can also expose now-playing information in-game.¹

The project builds on Simple Voice Chat’s current Minecraft 26.2-compatible Fabric release, so players only need the normal voice-chat client rather than a separate Jockey Radio mod.² The release also bundles ffmpeg and yt-dlp for managed hosts where shell access may not be available.

There is considerably more around scaling, early testing, platform-link safeguards, and setup in the full project page, which is worth reading if live audio inside a Minecraft server sounds interesting.

Author share note:

"We built Jockey Radio to make a Minecraft jukebox behave more like a real broadcast endpoint. You can attach a live station, push an RTMP feed from OBS or SavePoint, or play server-hosted audio and let nearby players hear it naturally in-game."

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow @gbtilabs's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/gbtilabs/

Advanced Pixel Editor Brings ImageMagick Editing Into WordPress

GBTI — Fri, 16 Oct 2026 22:51:06 +0000

By Rafael Minuesa, GBTI Network Supporter. Originally published on gbti.network.

Advanced Pixel Editor edits the images already in a WordPress Media Library without leaving the dashboard. Pick an image, move a slider, and ImageMagick renders the result on the server and sends
back a preview.

Image editing inside the WordPress Media Library

ImageMagick renders every adjustment on the server rather than in the browser, so the preview is the file that will be saved. The editor opens from the Media Library grid, the list view, or any attachment edit screen, and the image loads straight into it. JPEG, PNG, GIF and WebP are supported, up to 10 MB and 4096 by 4096 pixels by default.

Sigmoidal contrast and unsharp mask sharpening

Contrast uses a sigmoidal curve, which lifts midtones while holding the ends of the range, and sharpening is an unsharp mask with separate amount, radius and threshold controls. Both are the algorithms Photoshop and Lightroom use for the same two jobs. A draggable comparison slider sits over the preview, so an edit is judged against the original rather than from memory.

Crop, resize and DPI in the same panel

Crop uses an interactive overlay with free, 1:1, 4:3 and 16:9 presets. Resize holds the aspect ratio unless told otherwise, DPI is set alongside it, and rotation offers 45 and 90 degree steps as well as a free angle. Horizontal and vertical flips are one click each.

AI image editing from a text prompt

On WordPress 7.0 or newer, an image can be edited by describing the change, for example replacing a sky with a sunset. The request goes to whichever AI provider is already connected under Settings and
Connectors, an OpenAI or Google key, so the plugin adds no account and no bill of its own: the provider charges for its own usage. The result can be refined with another prompt, adjusted with the ordinary sliders, and saved like any other edit.

Automatic backup and one-click restore

Saving defaults to a new file, which leaves the original untouched. Replacing the original is the other option, and it writes an automatic backup first, so the original is one click away afterwards.
The editor is keyboard navigable, carries ARIA labels and screen reader support, and the comparison slider works on touch.

Advanced Pixel Editor Pro filters, watermarks and batches

The Pro add-on adds filters, watermarking and batch processing to the same editor. Filters cover brightness, saturation, hue, auto levels, sepia, vignette, Gaussian blur, noise reduction and emboss, each with the same live preview. Watermarking takes text or an image from the Media Library, with size, colour, opacity, rotation, a nine point position grid and tiling. Batch processing applies contrast, sharpness, brightness, saturation and hue across many images at once.

WordPress 5.6 and PHP 7.4 requirements

WordPress 5.6 or later and PHP 7.4 or later, tested to WordPress 7.1.3. The AI edit is the one feature with a higher floor, since it needs WordPress 7.0 and a connected provider. Version 3.8.0, released
under GPLv2 or later, with the source on GitHub.

The resource shared in this post is a new project published by GBTI Network member Rafael Minuesa.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow Rafael Minuesa's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/rafael-minuesa/

A WordPress Plugin for Detecting Distributed Bot Swarms

GBTI — Wed, 14 Oct 2026 20:51:36 +0000

By Rafael Minuesa, GBTI Network Supporter. Originally published on gbti.network.

Bot Storm Radar watches a WordPress site's traffic as a crowd and scores it as a whole, instead of judging one visitor at a time. A swarm of thousands of addresses that each make a single request looks perfectly ordinary to any per-IP rate limiter, because no individual address does anything wrong.

WordPress bot swarm detection

Every minute it counts the distinct addresses that visited, how many made exactly one request, how many loaded a stylesheet or a script the way a real browser does, and how widely the user-agent strings are spread. Those become a storm score from 0 to 100, compared against a baseline learned over seven days, and gated by traffic volume so a quiet site stays at 0.

Requests are classified as html, search, rest, xmlrpc, login, admin-ajax, asset or 404, with WooCommerce classes added when WooCommerce is present. Counters are kept in sliding one-minute and ten-minute windows keyed by address, IPv4 /24, IPv6 /48, user-agent hash and session, held in a persistent object cache where one exists, then APCu, then transients.

The site moves between four states (calm, warning, storm and cooling), and each transition is logged with the numbers that caused it. Addresses claiming to be Googlebot, Bingbot, Applebot or Yandex are checked by reverse and forward DNS before they are believed.

Blocking scanner probes and banning bad bots

Since version 0.2 the plugin also acts. A gate answers scanner probes with a 403: requests for .env and .git files, backups, database dumps, config and log files, and the admin paths of other applications. An address that keeps probing, or that keeps asking for pages that do not exist, is banned for a while.

Bans start in observe mode, where every trip is listed as a ban that would have happened and nobody is actually refused, so the list can be checked against real traffic before enforcement is switched on. Administrators, verified search bots, declared proxies and anything on the never-ban list are never banned.

Refusing bots before WordPress loads

On the GitHub build, a must-use plugin loads the gate before the rest of WordPress. It reads a small state file and touches neither WordPress nor the database, so a refused probe costs almost nothing. Turning on early protection goes one step further and writes an auto_prepend_file line into .user.ini under PHP-FPM, or a php_value line into .htaccess under Apache with mod_php, so the gate runs before PHP loads WordPress at all.

The plugin verifies with a test request that the gate really runs first, and if another auto_prepend_file is already set it reports the conflict rather than replacing it. Dropping an empty file named disabled into the data directory turns the gate off without wp-admin.

The wordpress.org plugin directory rule

The wordpress.org build does not carry the early gate. The plugin directory does not accept a plugin that writes PHP files, ships a must-use plugin or sets auto_prepend_file, so that build makes the same decisions later, on plugins_loaded, using rules kept in an autoloaded option. The refusals still happen; they cost more, because WordPress and the database have already loaded by then. Abuse reporting is GitHub-build only.

AbuseIPDB reporting from WordPress

With an AbuseIPDB key saved, probe bans can be reported, one per address per day and within a daily cap, covering probe trips only. Those reports are public. Provider reports look up a network's abuse contact through RIPEstat, sending only the address, and produce plain text to copy or hand to a mail client; the plugin never sends mail itself.

WordPress 6.0 and PHP 7.4 requirements

WordPress 6.0 or later and PHP 7.4 or later. A persistent object cache or APCu is strongly recommended, since without one the counters fall back to database transients. Released under GPL-2.0.

The resource shared in this post is a new project published by GBTI Network member Rafael Minuesa.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow Rafael Minuesa's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/rafael-minuesa/

Sondeas: A Browser Voice Workspace for Clinical Dictation

GBTI — Wed, 14 Oct 2026 00:55:21 +0000

By Jesus Iniesta, GBTI Network Supporter. Originally published on gbti.network.

Sondeas is a browser-based voice workspace. Record live or upload an existing file, and it returns an editable transcript you can translate, summarise, restructure or turn back into speech. It runs in Chrome, Edge and Safari on desktop and mobile, with nothing to install.

From recording to draft

It records in the browser, or takes an uploaded MP3, M4A, WAV, WebM or MP4 file, including multi-hour sessions. The transcript arrives punctuated, timestamped, split into paragraphs and separated by speaker, so it is readable before any editing. Rewrite presets then summarise it, cut filler, pull out action items or restructure it as an outline. Translation covers 100 languages, and text-to-speech generates audio back from the finished text. Labels organise transcripts by project, status or to-do list.

Built around clinical dictation

Sondeas is aimed at doctors and healthcare workers doing recurring dictation: record permitted, non-sensitive detail between appointments while the context is fresh, shape it into a draft, then review it. The site states plainly that Sondeas is not the EHR, and that the official record is still updated in an approved clinical system. Meeting notes, voice notes and research interviews are the other named use cases.

Privacy

Content is encrypted with a passphrase the account holder sets, rotates and revokes. Sondeas states that its staff cannot view user content, and that without the passphrase it cannot decrypt anything. Transcripts, audio and translations can be exported, downloaded or deleted at any time, and billing history carries timestamped credit records with downloadable receipts.

Plans

The Clinician Plan is 29 euros a month, with unlimited slow transcription while it is active and 1,500 premium minutes a month. Premium minutes cover fast transcription, translation, rewrites and speech generation. Pay-as-you-go minutes start at 2 dollars with no subscription, and purchased credits do not expire. A trial can be requested from the site.

Tools and guides

Alongside the workspace sit standalone pages for audio-to-text, voice-memo-to-text, text-to-speech and audio translation, plus guides for transcribing WhatsApp voice messages and iPhone voice memos, and a global keyboard shortcut for starting a recording.

The resource shared in this post is a new project published by GBTI Network member Jesus Iniesta.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow Jesus Iniesta's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/gsusi/

Magic Manta: Find and Verify a Company's Social Profiles

GBTI — Tue, 13 Oct 2026 00:36:54 +0000

Mike “Demo” DemopoulosFollow

By Mike Demo Demopoulos, GBTI Network Supporter. Originally published on gbti.network.

Magic Manta is a social radar scanner. Enter a company's domain and it finds that company's social media profiles across X, Threads, Instagram, LinkedIn, Bluesky, GitHub and more, then shows the evidence behind every handle it returns.

How a lookup works

It reads the site's Schema.org sameAs data and the links on its official pages, then falls back to targeted search for anything those do not declare. A profile declared through JSON-LD sameAs or an official-site DOM link gets a Verified badge, with its evidence source shown. A search fallback candidate stays Unverified unless it links back to the official domain. Every result carries a last-checked timestamp, and conflicts between official sources are flagged.

Beyond one domain

Batch lookups sweep several domains in a single run: five per batch on the Operative plan, twenty-five on Deep Recon.

For agents

The radar core is framework-agnostic, with three doorways into the same engine. The MCP server is live today, auth-gated, at 100 lookups per account per day. A REST endpoint and a React SDK are listed as coming soon, and there is no public REST API yet.

Accounts and plans

Single lookups need no account. An account unlocks saved lookups, lists, paid plans and the MCP server, with sign-in through Google, Apple, Microsoft, AgentID, or email and password. Single lookups are free. Operative is a one-time five dollars for an account, lookup history, saved lists and CSV or JSON export. Deep Recon is ten dollars a month, and Brand Command twenty dollars a month.

Hosting and carbon

The site runs on Lovable Cloud behind Cloudflare, which the Green Web Foundation verifies as green hosting, and 1.5 percent of purchases goes to Stripe Climate for carbon removal. A machine-readable disclosure sits at /carbon.txt.

The resource shared in this post is a new project published by GBTI Network member Mike Demo Demopoulos.

Are you a writer, musician, or product developer? We would love to support your work on the GBTI Network. For more information about how to join our community visit https://gbti.network

To follow Mike Demo Demopoulos's work more closely, consider joining our network and subscribing to them directly: https://gbti.network/members/mike-demo/