Thousands of frames land on the card after a wedding shoot, and Facet finds the keepers among the bursts, blinks, and near-duplicates. Facet scores every photo across nine dimensions, aesthetic quality, composition, face quality, eye sharpness, technical sharpness, color, exposure, subject saliency, and dynamic range, with category-specific weights for portraits, landscapes, wildlife, and 30-plus other content types. The culling darkroom groups bursts and visually similar frames, protects exposure brackets and panorama sequences, and flags closed eyes per face, so you keep the sharpest open-eyed frame and reject the rest in one pass. Face detection clusters into people you can rename, merge, and search, while albums, star ratings, favorites, and AI content tags organize what survives. Browse survivors through a mosaic gallery with semantic search by plain description, a timeline, a geotag map, and slideshow capsules. Statistics dashboards track your gear usage and metric correlations, and the weight editor learns from A/B comparisons so scores drift toward your own taste. Shared album links support client proofing with hearts and comments, phones auto-upload over the built-in WebDAV inbox, and ratings round-trip to Lightroom, darktable, and Immich through XMP sidecars. Everything runs locally on your own hardware: no cloud accounts, no subscriptions, no photo ever leaves the server. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

An order that waits days for a payment or a manager's approval survives crashes, restarts, and deployments in Cadence, a workflow engine that resumes each process exactly where it stopped. Developers write the order's logic as ordinary Go or Java code. The engine records every step in an event history and replays it to rebuild state on any worker. Activities send the emails, call the APIs, and write the database rows, each with its own retry policy, timeout, and heartbeat. A durable timer waits seconds or months without holding a thread, a cron schedule launches recurring runs, and a signal delivers the payment confirmation into the running workflow. Queries read its state without changing it, child workflows split big processes into reusable parts, and versioning allows code changes while older executions are still in flight. Domains separate teams and applications, task lists carry their own rate limits and priorities, global domains replicate across clusters and fail over between them, and archival retains closed histories. Cadence Web finds executions with SQL-like queries, lays out event history as a grouped table or a timeline, and exposes payloads, pending activities, and stack traces. It can start, reset, signal, cancel, or terminate runs, with batch actions for thousands of workflows at once. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

Convert paper archives, email attachments, and PDF records into an organized digital repository with Docspell, an open-source document management system that automates optical character recognition, metadata extraction, and full-text search. Users can ingest receipts and invoices through network scanners, monitored IMAP email mailboxes, mobile uploads, or drag-and-drop browser tools. The background processing pipeline executes optical character recognition via Tesseract, enhances image contrast with unpaper, and extracts text from Word documents and PDFs. Machine learning algorithms analyze document syntax to predict correspondents, suggest relevant organization tags, and identify due dates automatically. Team members can search their entire filing cabinet using complex Boolean queries and full-text indexing powered by Apache Solr or PostgreSQL. Administrators can configure multi-user collectives with isolated permissions, define custom metadata attributes, set up webhooks, and share time-limited document download links. Users can also merge multi-page scans, track processing job queues, and export curated document collections for tax filings or legal audits. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. GNU AGPL v3.0 licensed.

Your household's kitchen command center: shared grocery lists that sync in real-time across everyone's phones, a recipe book that imports from 500+ websites with one tap, a weekly meal planner that auto-generates shopping lists, and expense tracking that splits costs between roommates or family members. Native apps for Android, iOS, macOS, Windows, Linux, and web provide partial offline support so the list works even in store dead zones. The recipe-scrapers library with wild mode fallback handles imports from virtually any cooking site, extracting ingredients, steps, and images automatically — then converts recipe ingredients directly to shopping items with a single tap. Smart suggestions learn from your patterns over time, ordering items by your typical store route and recommending frequently purchased products. OpenID Connect authentication supports Google, Apple, Authelia, and custom providers. LLM-powered ingredient parsing uses configurable AI models for intelligent recipe text extraction. A Home Assistant integration enables voice-controlled list management via smart speakers. The REST API with a published Python client library allows third-party automation. Multi-language support covers 30+ languages through Weblate community translations. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

Document-generation code is the worst kind of code in your backlog - Carbone kills it. Its insight is separating design from data - templates are ordinary office documents (DOCX, ODT, XLSX, PPTX, HTML, even custom XML) built in LibreOffice, Microsoft Office, or Google Docs, with mustache-like markers such as {d.companyName} typed directly into the text. Send a template plus JSON from your existing APIs to the HTTP API, and Carbone returns the finished document - exported as-is or converted to PDF, XLSX, CSV, HTML, PNG, EPUB, and more via its integrated LibreOffice converter (Chromium and OnlyOffice engines are also supported for HTML-fidelity and office-format conversions). The template language goes well beyond substitution: loops over arrays render dynamic table rows, filters and aggregations run inside the document, and built-in formatters handle dates, numbers, currencies, timezones, and locales, with custom JavaScript formatters when needed. One template serves multiple languages through translation markers with auto-maintained translation files. The XML-agnostic engine means anything your document editor can design - pagination, headers, footers, nested tables, charts - survives generation intact, and Carbone guarantees no breaking changes in template syntax. Node.js-based, fast via multi-threaded LibreOffice conversion. The invoices, contracts, and reports your product owes its users become template edits, not sprints.

Define a DocType in Frappe and the framework automatically generates database tables, Python ORM classes, REST API endpoints, form views, list views, and role-based permission rules from that single schema definition. The meta-driven architecture stores schemas as database records, letting administrators add fields, reorder layouts, and modify validation rules through the browser-based Form Builder without redeployment. Server-side Python controllers handle business logic through hook-based lifecycle events while client-side JavaScript manages interactive form behavior, with WebSocket connections providing real-time updates across sessions. The built-in workflow engine defines document state machines with role-gated transitions, email notifications, and conditional action visibility. Pre-configured Desk views include list, form, report, tree, kanban, calendar, and dashboard layouts with drag-and-drop workspace customization. The report builder generates grid reports with configurable columns, filters, grouping, and chart visualizations from DocType data or custom SQL and Python scripts. Background job processing uses Redis queues for email delivery, data imports, and periodic tasks. Virtual DocTypes connect external databases to the framework's UI without data migration. Docker deployment via frappe_docker provides production-ready Compose configurations with MariaDB, Redis, Nginx, and worker containers. The foundation powering ERPNext, HRMS, CRM, and Helpdesk in production. Over 10,500 GitHub stars. MIT licensed.

With over 13,000 GitHub stars and active monthly releases, Mailcow delivers a production-ready mail server infrastructure through Docker Compose orchestration of twelve integrated containers. The stack pairs Postfix as the MTA handling SMTP traffic with Dovecot for IMAP and POP3 retrieval featuring the Flatcurve full-text search engine, while Rspamd provides machine-learning-powered spam filtering with Bayesian classification, DKIM signing, and ARC support. ClamAV scans attachments for malware including Office document macro analysis via Olefy, and SOGo serves as a responsive webmail client with CalDAV, CardDAV, and Microsoft ActiveSync synchronization. The admin panel manages multiple domains, per-user spam score thresholds, quarantine review, black and whitelists, temporary spam aliases, mailbox quotas, and TLS enforcement policies for both inbound and outbound messages. Security features include two-factor authentication via TOTP, WebAuthn, and Yubikey OTP, a Fail2ban-like Netfilter integration for brute-force protection, and automatic SPF, DKIM, and DMARC configuration. The ACME container handles Let's Encrypt certificate generation with DNS-01 challenge support, while Postfix TLS-Pol provides MTA-STS and DANE/TLSA resolution for encrypted transport. Built-in imapsync enables periodic migration from remote mailboxes, and the REST API allows programmatic management of all server resources. MariaDB stores user data, Redis caches configuration and sessions, and Unbound provides local DNSSEC-validating DNS resolution. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. GPL-3.0 licensed.

A PostgreSQL database becomes a production-grade GraphQL API the moment Hasura GraphQL Engine points at it: track tables and relationships - existing schemas included - and full query, mutation, and subscription types appear with where, order_by, limit, offset, and on_conflict arguments, no resolvers or boilerplate written. Its Haskell core compiles GraphQL to efficient SQL, and any query becomes a real-time live query with a single keyword, powering dashboards and collaborative UIs over standard GraphQL subscriptions. Authorization is where Hasura earns its enterprise reputation: role-based access control with row- and column-level permission policies driven by session variables from JWTs, auth webhooks, or headers - each role effectively sees its own GraphQL schema containing only what it may touch, integrating cleanly with Auth0, Firebase, or homegrown auth. Event triggers fire webhooks on inserts, updates, and deletes for asynchronous business logic; Actions extend the schema with custom REST handlers; remote schema stitching merges external GraphQL services into one endpoint; and auto-generated REST endpoints serve clients that skip GraphQL. A browser console handles data modeling and API exploration, the CLI manages migrations and metadata as code, and deployment is a single stateless Docker container beside Postgres.

Strength training, cardio, and body metrics land in one workout log on your own server, backed by an optional AI coach that explains its advice from your real numbers. Quick entry such as 100x8@9 logs weight, reps, and effort in one line, with a rest timer, plate-loading calculator, and warm-up ramp beside the set table. A weight picker offers only loads your equipment can build, and a readiness check-in adjusts the suggested load and says why. Templates for 5/3/1 BBB, GZCLP, nSuns, PPL, and Upper/Lower run as written and edit like any custom program. Progress pages chart estimated 1RM, weekly volume per muscle group against MEV and MRV bands, a muscle heat map, bodyweight goals, and a training calendar. Garmin FIT, TCX, and GPX imports need no cloud account and add heart-rate charts and aerobic decoupling to each run or ride. The coach writes a weekly debrief, chats in a streaming conversation, and drafts a full program from a one-sentence goal, using your own Anthropic or OpenRouter key. History imports from Strong or Hevy CSV files with a dry-run preview and exports to CSV or TCX, and an MCP connector lets ChatGPT or other agents analyse it. The app installs as an offline-capable PWA, speaks English, French, Russian, and German, and keeps every user's data separate. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Galene is a self-hosted video conferencing system that runs full meetings, lectures, and tutorials in the browser while asking remarkably little of the server it runs on. Every group gets its own page with video tiles, a user list, and a chat pane, and participants join from desktop or mobile browsers with nothing to install. Presenters share a screen or individual windows, stream a video or audio file straight from disk, and switch to blackboard mode when a slide needs resolution more than frame rate. Moderators can warn or remove participants and lock the group, raise-hand status keeps question queues orderly, and anyone can blur their background or exchange files directly with peers. Groups are simple configuration files: set public or private rooms, passwords or login tokens, client limits, and expiry windows, then create users with moderator or presenter privileges using the galenectl command. Automatic subgroups spin up identical breakout rooms on demand, the pattern its author uses to supervise forty students working in pairs during computer science practicals. Calls stay smooth on poor networks because the server adapts video quality per participant, recovers lost packets, and relays media through its built-in TURN server when firewalls block direct paths. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Circulation desk, cataloging workbench, acquisitions office, serials room, and public catalog all share one roof in Koha, an open-source library system that scales across school libraries, branch networks, and consortia. Staff search the catalog with facets for availability, authors, holding libraries, and item types, then open a record to see its cover, subjects, classification numbers, and every copy with its call number, status, and barcode. Catalogers edit MARC 21 or UNIMARC records field by field, copy records in via Z39.50 and SRU searches, and link authorities automatically. At the desk, staff check out, check in, and renew items, work with holds, bookings, and claims, and see overdue warnings and outstanding charges beside each patron. Circulation and fine rules vary by library, patron category, and item type. Acquisitions handles vendors, baskets, budgets, and funds. Serials track subscriptions and issue receiving. A guided report wizard, SQL reports, and statistics wizards summarize circulation and collections. Patrons search the public catalog, place holds, save lists, tag and rate titles, and suggest purchases. Staff tools design notices and slips, print patron cards, set holiday calendars, and import or batch-edit patrons. SIP2 serves self-checkout kiosks, OAI-PMH sets share records, and a REST API exposes patrons, holds, checkouts, and biblios. Lending continues during outages with offline circulation. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. GPL-3.0 licensed.

Friendica connects to more federation protocols than any other single platform: ActivityPub reaches Mastodon, Pleroma, Misskey, Lemmy, Pixelfed, and PeerTube; the Diaspora protocol bridges that network; and OStatus provides legacy compatibility. But protocol reach is just the foundation. Where most fediverse platforms stop at microblogging, Friendica offers long-form posts, photo albums, event calendars with RSVP, group discussions, and rich multimedia content. BlueSky and Tumblr connectors extend your reach into non-federated networks, while RSS import pulls content from any blog or news source into your timeline. Privacy controls let you scope each post to specific contacts, groups, or public visibility with per-post granularity rather than account-wide settings. Multiple themes customize the interface from Facebook-like layouts to minimal designs, and Fedilab provides native mobile access. Under the hood, the PHP application with MariaDB runs as either an Apache or FPM Docker image with Redis caching and background cron workers handling federation delivery, media processing, and notification dispatch. One of the oldest fediverse projects (1,650+ stars, 160+ contributors since 2010), Friendica has weathered every protocol evolution and platform migration the decentralized web has thrown at it, making it one of the most battle-tested social networking platforms available.

In continuous open-source development since 2009, Etherpad is the original really-real-time collaborative editor - used by Wikimedia, governments, EU public-sector institutions, and tens of thousands of self-hosters. Its core idea is visible authorship: every keystroke is attributed with author colors, every revision preserved, and the timeslider lets you scrub through a document's entire history character by character. Multiple people type into the same pad and see each other's changes instantly - it scales to thousands of simultaneous editors per pad. The base install is deliberately lightweight; capability comes from roughly 290 plugins installable from the admin web UI: comments, images, tables, drawing, video chat via WebRTC, math rendering, code highlighting, and authentication via OAuth, LDAP, or OpenID. AI is pointedly a plugin, not a default - you choose the model and infrastructure, or never turn it on. There is no telemetry. For integrators, an HTTP API (with OpenAPI definitions at /api/openapi.json) manages pads, users, and groups for embedding in your own applications, and the ueberDB abstraction layer supports PostgreSQL, MySQL, Redis, MongoDB, and SQLite backends. Full data export is built in, the format is open, it is translated into 105 languages, and it runs on anything from a Raspberry Pi to a server farm. Apache 2.0 licensed, Node.js based.

Securo is a privacy-first personal finance platform that keeps all your financial data in your own PostgreSQL instance while connecting to 2,500+ banks across three continents for automated transaction sync. The FastAPI backend with React/TypeScript frontend deploys via Docker Compose on ports 3000 and 8000, with Redis-backed Celery workers handling background synchronization. Bank connections span approximately 2,500 European PSD2 institutions through Enable Banking across 29 countries, US and international banks via SimpleFIN's open protocol, and Brazilian banks through Pluggy, all refreshing automatically every four hours. Multi-account management tracks checking, savings, credit cards, and investments with running balances. The auto-categorization rules engine applies pattern-matching across merchants and amounts. Goal tracking monitors savings targets with progress visualization, asset management handles stocks and custom instruments with configurable valuation growth rules, and reports generate Net Worth and Income vs Expenses charts with category sparklines. Optional AI Agents provide LLM chat with tool-use over your financial data through MCP, supporting OpenAI, Anthropic, Ollama, and compatible providers with per-agent RAG knowledge bases accessible via a global command palette. Multi-user support includes OIDC login, passkey authentication, TOTP two-factor, privacy mode, and multi-currency FX conversion via Open Exchange Rates. 1,500+ stars since March 2026. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

Google's search results without Google's surveillance: Whoogle is a self-hosted proxy that strips the tracking and keeps the results. Your query goes from browser to your Whoogle instance, which fetches results from Google with a randomly generated User Agent and strips everything hostile before returning them: no ads or sponsored content, no third-party JavaScript or cookies, no AMP links, no URL tracking tags like utm_source, no referrer header - and Google sees your server's IP, never yours. Unlike metasearch engines that blend sources, Whoogle proxies Google exclusively, so result quality is exactly what you'd get logged out and incognito, minus the noise. A lightweight Flask app configured entirely through environment variables, it supports DuckDuckGo-style bang shortcuts, autocomplete suggestions, safe search, per-country and per-language filtering, site blocklists, and automatic rewriting of social links to privacy front-ends like Nitter and Invidious. Privacy hardening goes further: built-in Tor routing makes Google see an exit node instead of your server, HTTP/SOCKS proxy support covers other setups, and POST-based queries keep search terms out of logs. Light, dark, and fully custom CSS themes plus browser search-engine registration make it a drop-in default on desktop and mobile. Stateless, tiny, and trivial to run.

Upload a report, ask a "what if" question in plain language, and MiroFish builds a small simulated society that plays out the answer before you commit to a decision. It reads your seed documents, whether news, policy drafts, financial signals, or a novel, and extracts entities and relationships into a GraphRAG knowledge graph. Entities from the graph become agents with a profession, a biography, and interest tags, and the app estimates simulation rounds from the simulated time span, with a custom mode for shorter runs. Agents post and react on two simulated platforms at once, a feed-style plaza and a topic community, while a live monitor tracks rounds, elapsed simulated time, and actions. A ReportAgent then plans an outline and writes the prediction report section by section with the InsightForge, PanoramaSearch, QuickSearch, and InterviewSubAgent tools. Afterward you can keep questioning the ReportAgent, chat with any individual agent, or send a survey question to a selected group and read their answers. Typical uses include rehearsing public relations responses, testing policy drafts against likely reactions, and exploring alternate endings for a story. The interface switches between English and Chinese, and you bring your own model: any OpenAI SDK compatible LLM endpoint works, Zep Cloud stores agent memory, and the project suggests starting under 40 rounds because runs consume many tokens. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

The inverter app tells you the sun shone; SOLECTRUS tells you what that sunshine was worth. It is a dashboard for your whole photovoltaic system that balances every watt: what the panels produced, what the house consumed, what the battery stored, and what crossed the grid meter in either direction, for any period you care to name. Up to five generators, from the main roof array to a balcony power plant, appear individually or summed, and up to twenty custom consumers, wallbox included, get their own accounting. The Power-Splitter shows how much green electricity actually landed in the car, the heat pump, or the washing machine, and what each load truly cost once lost feed-in revenue counts. Enter your electricity prices and feed-in tariff, including their changes over the years, and it computes savings, self-consumption costs, return, and the day the system has paid for itself from measured data rather than the sales brochure. Forecasts from pvnode, Solcast, or Forecast.Solar look up to seven days ahead using your location, orientation, tilt, and peak capacity. Heat pump owners see generated heat and coefficient of performance beside the power draw. Everything lands in InfluxDB and stays there, with daily aggregation keeping multi-year views fast, trends against last month or last year, and a Top-10 of your best and worst days. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.

Pingvin Share X is your own WeTransfer, running on your server, with none of the transfer limits or expiring-link anxiety of the free tiers. Drop in a file of any size, get a link, and decide exactly how it behaves: an expiration date after which it vanishes, a password gate, a cap on how many visitors may open it, or all three. Email the link straight to recipients from the share dialog, and get notified when they download. Reverse shares flip the flow around: send someone a link and they upload files to you through it, which is perfect for collecting client deliverables, application documents, or the family photo archive without a single email attachment thread. Attachments can be scanned with ClamAV before anyone touches them, and storage lives on your disks or in S3-compatible buckets. Accounts connect through OIDC or LDAP single sign-on for teams, and an extensive admin panel tunes everything from password policy to upload chunk sizes. Registration rules decide whether your instance is a private family tool or an open drop point for your whole crew. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. BSD-2-Clause licensed.