One self-hosted workspace holds your team's chat, tasks, notes, and calendar, and Deft lets AI agents work in it beside your people. People get real-time chat with spaces, direct messages, threads, mentions, reactions, files, and presence. Tasks open as Board, Table, Timeline, or Calendar views, with dependencies, subtasks, recurrence, and bulk actions. Notes, decisions, and channel memory become durable knowledge, with graph views and links back to the source conversation. Native calendar events sit beside read-only ICS subscriptions. Agent employees can join channels and receive task assignments. Tag the built-in assistant Defty to draft a task, answer a workspace question, or take a governed action, then review the proposal in an approval card before anything changes. Personal connections let Codex, Claude, ChatGPT, and other streamable HTTP MCP clients act as the signed-in user within revocable scopes. Approval tiers, trust levels, token scopes, revocation, activity history, and action receipts record who or what acted and what changed. OpenAI, Anthropic, OpenRouter, OpenAI-compatible endpoints, and local Ollama-style models are supported providers. Chat, tasks, notes, and calendars keep working with no AI key. Modules add company-specific records, relationships, and views, with a bundled Contacts module as the first example. 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-only licensed.

Most RSS readers hand you a pile of headlines and make you do the reading; FeedMe reads first and hands you the pile already digested. It aggregates every feed you configure into one clean, ranked stream, and for each article an LLM writes a short summary you can scan in seconds, with the full original text one tab away when a story earns a deeper look. Sources organize into categories like Tech News, Finance News, Personal Blogs, Company Blogs, Open Source, Research, and Podcasts, and a global search finds stories across all of them. The reader never sleeps on stale content: a built-in cron refreshes feeds and regenerates summaries on your schedule (every three hours by default in Docker), so the page you open in the morning already covers the night. Summaries answer to you, not a vendor: point FeedMe at any OpenAI-compatible endpoint and tune the prompt, input truncation, temperature, and token budget, with output in the languages you choose. There are no accounts, no forced logins, and no app to install; the responsive page works on phone and desktop in light or dark themes. Because the whole application is a static site, it stays fast and cheap to run, and the same build can live on GitHub Pages or Vercel if you ever want a public edition next to your private one. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

ThoughtDAG turns a long AI conversation from a scroll you cannot rearrange into a map you can edit, branch, and reconnect. Every question and answer becomes a node on an infinite canvas, and the wires between nodes are not decoration: they decide exactly what context the model sees next. Follow a promising thread, park a side question on its own branch, or cut a wire when an early assumption starts poisoning later answers, all without deleting the work that led there. Zoom out to read the shape of an investigation, then zoom back in to full answers, attachments, and highlighted passages. Drop in documents and files as material nodes, ask questions from them, and preview precisely what will be sent before a model runs. Past coding agent sessions can be opened as read-only graph mirrors, so a discussion that started in a terminal can be inspected, cited, and continued on the canvas. Memory lives in readable dossiers that collect decisions and open questions per topic, with sources attached, and relevant dossiers join new questions when you ask. Export a backup, a Markdown context chain, or a thought map that shares the structure of your thinking without dumping the whole transcript. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Metadata Remote is a browser-based audio tag editor designed specifically for headless music servers, enabling collectors to fix ID3 tags, embed high-resolution cover art, and match releases against MusicBrainz without moving files off their NAS. Navidrome, Jellyfin, and Plex administrators can browse mounted directories directly from any browser, inspect standard and extended metadata fields, and apply single-file or bulk folder-wide updates in one click. Clicking an empty or incomplete field activates the smart inference engine, which analyzes filenames, directory hierarchy, and sibling tracks alongside live MusicBrainz queries to present candidate values ranked by confidence percentage, including specialized matching rules for classical works, multi-artist compilations, and live bootlegs. Built-in audio streaming lets you audition MP3, FLAC, OGG, OPUS, M4A, M4B, WMA, WAV, and WavPack files inside the interface to verify mysterious tracks before renaming files or folders inline. A dedicated long-form text editor automatically opens for lyrics and liner notes exceeding 100 characters, while the album art manager previews image dimensions, bulk-embeds covers across entire albums, and repairs corrupted artwork headers automatically. Every single-track edit, custom field creation, and batch folder operation is tracked in a 1,000-step undo and redo history with full keyboard-only navigation. 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.

Instant board updates, side-by-side card multitasking, and a five-level task hierarchy make 4ga Boards a fast, clutter-free Kanban workspace for teams tired of sluggish project trackers. Work organizes cleanly from top-level projects down into boards, collapsible lists, cards, and granular subtask checklists. You can switch between classic drag-and-drop Kanban columns and compact to-do list views with a single click. While editing a card description in the rich Markdown editor with colored text, you can still scroll, filter, and rearrange other cards on the board without losing unsaved changes. Every move, label change, due date update, or comment syncs live across connected browsers without page reloads. Collapsible lists and a retractable sidebar free up wide-screen horizontal space during sprint reviews. Teams migrating existing workflows can import boards directly from Trello JSON exports or CSV files and export backups on demand. Cards support file attachments, custom cover images, multi-member assignments, and @mentions with instant activity notifications. Authentication integrates with Google, GitHub, Microsoft, and custom OIDC single sign-on providers alongside role-based board privacy. The interface includes dark mode, full keyboard shortcuts, and localization across 16 languages. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Wag is an authentication gateway for WireGuard that lets each device reach only the routes it should, and asks for a second factor before the sensitive ones open. Admins write access policies for everyone, groups, or single users, marking subnets, hosts, ports, and protocols as MFA-required, always allowed, or denied, and the most specific prefix wins, so one allowed address can sit inside a protected range. Devices join with a registration token, and the public endpoint answers with a finished WireGuard profile. Users then confirm a TOTP code, a WebAuthn security key, an OIDC single sign-on login that maps identity provider groups into Wag groups, or a PAM prompt, and repeated failures lock the account. Session lifetime and idle timeouts decide when they must prove themselves again. The browser portal lists users, devices, rules, registration tokens, and webhooks with real-time notifications, while a command line locks devices, resets MFA, and manages administrators. Webhooks turn JSON from an HR or identity system into new tokens or removed users and devices, and clients can fetch the list of guarded routes from a small API. Remote staff, contractors limited to one host, and homelabs are typical uses. When a server fails, the VPN remains reachable through clustering with witness nodes or an external etcd, and certificates arrive through ACME with HTTP-01 or Cloudflare DNS-01. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. BSD-3-Clause licensed.

Meeting-app screen sharing was built for slide decks, and it shows when two engineers chase a bug across a codebase together; Hopp rebuilds screen sharing for exactly that job. Click a teammate and their screen opens in under a second: no meeting links, no calendar invites, no "can you see my screen?" ritual. Both sides get real keyboard and mouse control with sub-100ms latency, so guests type, click, and navigate instead of dictating instructions. Rooms hold up to 10 people for mob programming, code reviews, and onboarding, all with shared cursors so everyone can point at the line they mean. When words fail, draw or type annotations directly on the shared screen. The stream defaults to 4K, so small fonts and subtle diffs stay readable, and noise-cancelling audio keeps voices clear with no separate call app. You share any application, not just a browser tab, so debugging in Grafana or reviewing a design works as well as editing code. App Veil hides the windows you pick, such as password managers or chat, from the shared screen. Slack starts sessions with one /hopp command, and custom keyboard shortcuts keep frequent actions in flow. The server is yours: a Docker Compose stack with PostgreSQL, Redis, LiveKit, and Caddy auto-TLS serves the web dashboard and media, and the desktop app points at your backend, not a vendor cloud. 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.

Invoicerr nails the exact billing sequence freelancers actually use: draft a quote, send it to the client, collect a legally-binding electronic signature via a secure browser link, convert the approved quote into an invoice, and generate a branded PDF — no DocuSign account, no external signing service, no enterprise complexity. The React frontend with NestJS backend uses Prisma ORM with your choice of SQLite for instant local setup or PostgreSQL for production workloads. Quotes and invoices are first-class objects with independent status tracking — unread, sent, viewed, signed for quotes; paid for invoices — not a single form with a label swap. The e-signature system generates secure-token signing links that clients open in their browser to review and sign documents without needing any account or external service. PDF generation renders quotes, invoices, and receipts with your company logo, name, VAT number, colors, and custom email templates so documents reflect your brand identity rather than the software's defaults. Authentication supports JWT tokens or OIDC providers stored in cookies. The REST API backend enables future integrations with mobile and desktop clients. A plugin system with community-made features and outgoing webhooks extends functionality without forking core code. International-friendly with customizable currencies. Deploy with Docker Compose using the prebuilt image from GitHub Container Registry. 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.

Every homelab deserves a control center that looks as good as the services it manages. Lab Dash gives you a customizable, drag-and-drop dashboard where every tile earns its spot. Add shortcut widgets linking to your Plex, Grafana, Nextcloud, or any service with automatic health check indicators showing green or red status at a glance. The system monitoring widget displays real-time CPU usage, memory consumption, network throughput, disk utilization, and temperature readings by mounting the host /sys directory as a read-only volume. Integrate Pi-hole stats, qBittorrent download progress, and media server widgets for Jellyfin, Plex, or Emby directly on your grid. Multiple pages let you organize services by category, and the search bar finds any tile instantly. Customize the background image from a local upload or URL, set a custom title and tab name, and choose your preferred search provider including self-hosted options like SearXNG. All sensitive configuration data is encrypted locally with AES-256-CBC, and only administrator accounts can modify the layout. Install Lab Dash as a Progressive Web App on desktop, phone, or tablet for a native app experience with offline capability. Configurations export as backup files for easy migration between instances. On RepoCloud, deploy Lab Dash on a dedicated VPS with root SSH access and Docker socket connectivity under the GPL-3.0 license.

Snitch is a self-hosted network monitor and outbound firewall that shows every connection your server makes, which process opened it, and where in the world it goes, then lets you block it on the spot. A live force graph puts your machine at the center with each remote host as a node, while a world map plots geolocated destinations and a sliding timeline keeps the last day of traffic searchable. Every alert is actionable: block the IP, kill or suspend the process, or write a rule that firewalls matched hosts automatically by address, domain, port, country, or ASN. Detection covers port scans, beaconing, DNS tunneling, ICMP tunnels, cleartext ports, and self-signed or mismatched TLS certificates, and an isolation-forest model learns your normal flows and flags outliers. A lockdown switch cuts all public outbound traffic while the LAN stays up, for the moment something looks wrong. Findings can stream to Wazuh, Splunk, or Elastic as syslog and ECS events, or land in Discord, Slack, and Telegram as webhook alerts. Everything stays on your machine: history lives in a local database, geolocation works fully offline, and the interface speaks eight languages. 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.

Webtor turns any magnet link or .torrent file into instant in-browser streaming and direct downloads on your own server, with no desktop torrent client and nothing to install for viewers. Drop in a magnet, an infohash, or a torrent file and the file list appears in seconds; pick a video and it plays in the built-in player while the pieces are still arriving, or take a plain HTTP download link for any single file instead. Whole torrents come down as ZIP or TAR archives generated on the fly with the folder structure preserved, and formats from MKV and MP4 to FLAC and MP3 stream without a separate transcoding step in most cases. Torrents you keep go into a personal library that recognizes movies and series automatically and enriches them with posters and descriptions, and the same library can mount as a network drive over WebDAV or S3 and play on a TV through the Stremio addon. Subtitles are matched from OpenSubtitles inside the player, uploaded tracks are supported, and optional AI translation renders them in your language; a master password locks the instance, the player embeds into other sites through the JavaScript SDK, and a REST API drives everything from your own code. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Running a single AI agent is straightforward; running a team of specialized agents that delegate tasks, share memory, and coordinate through structured workflows requires an orchestration layer, and that is exactly what SwarmClaw provides. Define a hierarchy of agents in an org chart where a Coordinator (your CEO agent) delegates research tasks to a Researcher, coding tasks to a Developer, and design tasks to a Designer, each configured with its own LLM provider, tool permissions, and skill set. The Task Board presents a Kanban view of all work items across Backlog, Queued, Running, and Completed columns, with each task card showing its assigned agent, tags, due dates, and approval gates that pause execution until a human reviews and approves. Agents execute work using built-in tools for file operations, shell commands, browser automation, and persistent memory, plus any MCP server you connect via stdio, SSE, or streamable HTTP transport. Durable structured sessions support branching logic, repeat loops, parallel branches with explicit joins, and restart-safe run state that survives crashes without losing progress. Over 23 LLM providers ship built-in: Claude Code CLI, OpenAI, Anthropic, Google Gemini, DeepSeek, Groq, Mistral, xAI Grok, Fireworks, Ollama, and more. Connectors push messages to Discord, Slack, and Telegram, while cron schedules and webhooks trigger agent runs automatically. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

After the bills are covered, how much is actually safe to spend? monetr organizes an entire budgeting application around that one question. Inspired by the defunct Simple bank, it budgets paycheck by paycheck rather than month by month. Funding schedules encode when you get paid - including multiple schedules per account and a weekend-exclusion option for deposits that land early - and every expense or savings goal is tied to one. monetr then calculates how much of each recurring obligation (rent, car payment, subscriptions, on any repeat interval) to set aside from each paycheck, so a large bill never has to come out of a single check. What remains after allocations is surfaced as Free-To-Use, and a forecasting timeline projects contributions, due dates, and leftover funds forward so low-balance periods are visible before they happen. Transactions arrive either as OFX uploads from your bank or automatically through a Plaid connection using your own API credentials, keeping balances and transactions synced without manual entry. The app is mobile-friendly and installable as a PWA on desktop or phone. Self-hosting via Docker Compose is completely free, with all source code public and your financial data on your own hardware.

Apache OpenMeetings puts the whole meeting in the browser: cameras, a shared whiteboard, the agenda document, and recording controls, all running on your own server. Participants join rooms with audio and video, audio only, or video only, choose camera resolutions and devices, and present by sharing a screen or a single window at selectable quality. Moderators control the floor person by person: who may draw on the whiteboard, present, share or record the screen, mute others, or speak while everyone else stays muted. Multiple whiteboards hold drawings, text, clip art, and imported documents in PDF, office, and image formats, rescaling to fit every participant's screen, and boards can be saved into the drag-and-drop file explorer for the next session. Every meeting can be recorded with all audio streams exactly as heard, then downloaded as MP4 or played back in the integrated player, and the built-in calendar schedules rooms and sends invitations. Breakout rooms split classes and workshops into groups, polls gather decisions live, private and group chat runs beside the video, and telephone users can dial into rooms through the Asterisk SIP integration. 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.

Foreign-language feeds are where good information usually goes to be ignored, and RSSBox fixes that by translating, filtering, and merging RSS sources, then republishing them as clean feeds your existing reader can subscribe to. Add a feed, pick a translation engine, and RSSBox republishes it with titles, full articles, or summaries translated, optionally showing both languages side by side. Each feed can use a different engine, from DeepL to any OpenAI-compatible model endpoint, and the admin shows translation status plus the tokens or characters each source has consumed. Update schedules are set per feed, so a fast-moving news source refreshes hourly while a quiet personal blog checks in once a day. Keyword filters quietly drop the recurring topics you never open, and AI filtering goes further by judging items against what you actually read. Tag several related sources and RSSBox merges them into a single unified feed, translations and filters applied, ready to follow as one stream. Full-text fetching pulls complete article bodies from feeds that only publish excerpts, so the translation covers the whole story rather than the teaser paragraph. Your reader app never changes: it simply subscribes to the translated feed URL like any other source. 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.

Centralizing BorgBackup usually means hand-creating Unix users, wiring SSH keys, and remembering which machine stopped backing up three weeks ago; BorgWarehouse replaces that terminal work with a dashboard built for exactly this job. Add a repository and it provisions the dedicated user, authorized key, and storage quota for you, then hands back ready-to-paste client commands so a new machine is backing up in minutes. Every repository carries a freshness threshold: when no modification lands within the window you set, its status dot turns red and an alert goes out by SMTP email, through Apprise to Discord, Telegram, Slack, Gotify, or ntfy, or as a signed webhook POST to tools like n8n and Uptime Kuma, rate-limited to one alert per repository per day. Append-only mode is enforced on the server, so even a fully compromised client can add archives but never destroy existing ones, which is the property that keeps ransomware from eating your backup history. The monitoring view tracks storage consumption against each quota, a setup wizard walks new clients through onboarding (with a LAN variant for local transfers), and sign-in can delegate to GitHub, Google, Microsoft, GitLab, or any OIDC provider. A versioned REST API automates provisioning, and your passphrase never reaches the server: backups stay end-to-end encrypted on the clients. 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.

Traffic Source answers the three questions every site owner actually asks: who showed up, where did they come from, and did they pay. The live dashboard counts visitors in real time alongside pageviews, sessions, bounce rate, and session duration, with no cookie banner required because tracking uses no cookies at all. Referrer and UTM breakdowns show exactly which campaigns, newsletters, and posts earn their traffic, while country and city data arrives through proxy headers. Visitor journeys replay every page a person viewed before converting, so you can see the path from first click to checkout instead of guessing. Connect Stripe once and payments are matched to the sessions that produced them automatically, with no webhook plumbing to maintain. The affiliate system lets you create partners with custom commission rates and shareable referral links, and each affiliate gets a public dashboard showing their own clicks and earnings. Search Console data is pulled into the same screen, where a keyword explorer shows the pages, countries, and devices behind any query. One tiny tracking script covers all your sites, including single-page apps. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

A free, polished web client for Subsonic-compatible music servers: Substreamer is the browser-based frontend you point at your existing streaming backend to play your own library from anywhere. It speaks the Subsonic API (v1.13 and higher), which makes it compatible with the whole ecosystem that has grown around that protocol: the original Subsonic server, its forks Airsonic and Madsonic, and modern implementations like Navidrome and Ampache. That decoupling is the point - your music files, transcoding, and library indexing live on whichever server you prefer, while Substreamer provides the listening experience: browse by artist, album, and genre, build and manage playlists, search your collection, and stream on demand. This RepoCloud deployment runs the containerized web edition, so the same interface is available from any browser without installing a native app, and it pairs with the Substreamer mobile apps that made the client popular. For anyone assembling a self-hosted Spotify replacement - typically Navidrome for the backend plus a good client - Substreamer fills the client half with a clean, familiar player UI. Because it is a stateless client, the container is lightweight and low-maintenance: connect it to your server's URL and credentials, and your entire collection is streaming in minutes, with no subscription and no catalog that can disappear.