T3 Code puts every coding agent you already pay for behind one control surface, so Claude Code, Codex, Cursor, Grok Build, OpenCode, and Antigravity all run on your server while you direct them from your phone, browser, or desktop. Start a thread, choose the provider and model that fit the job, and watch the agent work in a shared conversation instead of a terminal you have to babysit. Permission modes decide when an agent must stop and ask, and its approvals, questions, and plan reviews arrive in the same thread wherever you happen to be connected from. When the work lands, review it file by file in the built-in diff view, then commit, push, and open pull requests through the Git and GitHub integrations without leaving the app. A real terminal sits beside the chat, and projects, settings, and scheduled tasks stay attached to the machine where the code actually runs. Install it as a background service and reach it from anywhere through T3 Connect, a Tailscale or LAN pairing, or the hosted web client, with a pairing code guarding local access. Outside agents can also drive threads over MCP, launching, messaging, and waiting on work across machines you have linked together. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Scoring 64.4 on the WebBench benchmark — state-of-the-art among browser automation platforms — Skyvern replaces brittle XPath-based scripts with Vision LLM reasoning that adapts when websites change their layouts. The platform extends Playwright with AI-powered page methods including page.act(), page.extract(), and page.validate() that accept natural language prompts while still supporting traditional CSS selectors as fallback. The drag-and-drop Workflow Studio offers 17+ block types including navigation, extraction, login, loops, conditionals, code blocks, file download, and file upload — enabling non-technical users to build complex multi-step automations without writing code. Self-hosted deployments support bring-your-own-LLM with OpenAI, Anthropic, Gemini, and Ollama, while the multi-engine architecture allows swapping between Skyvern 2.0, OpenAI CUA, Anthropic CUA, or UI-TARS per task with a single parameter. Built-in infrastructure handles persistent browser sessions preserving cookies and localStorage across runs, automatic CAPTCHA solving for reCAPTCHA and hCaptcha, anti-bot bypass for Cloudflare and DataDome, residential proxy rotation across 20+ countries, and a credential vault integrating with Bitwarden and 1Password for secure 2FA management. Real-time session livestreaming via WebRTC enables visual debugging, while step-by-step action logs with screenshots and full LLM diagnostic traces provide production observability. The MCP server integration exposes Skyvern as a tool for Claude, Cursor, Windsurf, and any MCP-compatible AI agent. Connect to 6,000+ apps through Zapier, Make.com, or self-hosted N8N workflows. Deploy via Docker Compose or pip install with a two-command setup. 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.

Cognee gives AI agents persistent long-term memory that survives across sessions, replacing the traditional stack of separate graph, vector, and session databases with a unified engine running on a single PostgreSQL instance. The memory-native API exposes four verbs (remember, recall, forget, and improve) enabling agents to persist context, retrieve cited answers, prune outdated knowledge, and self-improve from feedback. Under the hood, Cognee combines pgvector embeddings with a PostgreSQL-native graph store and cognitive-science-grounded ontology generation, delivering hybrid retrieval that fuses semantic similarity, structural graph traversal, and lexical search in a single query. Integrations span Claude Code, Cursor, LangGraph, OpenAI Agents, and any MCP-compatible client through a dedicated MCP server on port 8001, while the Python and TypeScript SDKs provide direct programmatic access. The platform supports swappable backends including Neo4j, FalkorDB, Qdrant, ChromaDB, Weaviate, Milvus, and LanceDB for teams with existing infrastructure. Built-in OpenTelemetry tracing, an experimental dashboard with knowledge graph visualization, multi-tenant user isolation, and audit trails ensure production readiness. Deploy via Docker Compose with optional profiles for PostgreSQL, Neo4j, Redis, and the web frontend. Reached v1.0 in April 2026 with 30,000+ stars. 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.

Foreman is the control room for a server fleet, taking machines from blank hardware or an empty VM to configured, monitored hosts through one web interface. Create a host entry and it network-boots by PXE or UEFI HTTP, or starts from a discovery image that phones home its hardware inventory so unknown machines can be claimed and built by rule. Installation runs from templates for Kickstart, Preseed, AutoYaST, Agama, and cloud-init, while host groups bundle the operating system, partition layout, and parameters that make every build of a role identical. The same console provisions onto bare metal, libvirt, VMware, oVirt, Amazon EC2, Google Compute Engine, and OpenStack, because each provider plugs in as a compute resource with its own images and networks. After installation, Puppet classes and parameters, Ansible roles with recurring runs, or Salt states keep configuration enforced, and the reports and facts those tools send back fill dashboards showing which hosts are current, which drifted, and which are erroring. Smart Proxy services carry DHCP, DNS, TFTP, and configuration enforcement into each subnet or datacenter segment, remote execution runs jobs across the fleet, and the Katello plugin adds content and patch management on top. Hammer gives the whole API a command-line face, and role-based access with LDAP or FreeIPA sign-in keeps large teams in their lanes. 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.

Sketch a wireframe, get working code: Draw a UI turns hand-drawn layouts into web interfaces. It pairs the open-source tldraw canvas with an OpenAI vision model: you sketch a layout - boxes, labels, buttons, arrows, whatever communicates the idea - select the drawing, and click Make Real. The app snapshots your selection as a PNG, sends it to the vision API with instructions to return a single HTML file styled with Tailwind CSS, and renders the result in an iframe directly on the canvas next to your sketch. The loop is iterative: annotate the generated prototype or redraw parts of it, select both the sketch and the previous result, and generate again - the model receives the earlier HTML as context and produces an updated version. Built by Figma engineer Sawyer Hood as one of the first viral GPT-4 Vision demos and the basis for tldraw's "Make Real", it is a Next.js app that runs against your own OpenAI API key. Self-hosting matters here: the upstream demo ships without authentication, so a private deployment keeps your API key from being drained by strangers. MIT-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.

Point OBS or a hardware encoder at Restreamer's built-in RTMP or SRT ingest and it serves your website while rebroadcasting to YouTube Live, Twitch, Facebook, Vimeo, TikTok, LinkedIn, PeerTube, and anything else that accepts RTMP, SRT, or HLS - a complete self-hosted live-streaming server from datarhei. That multistreaming normally costs a monthly Restream.io subscription; here it's one FFmpeg process per destination on your own hardware. The web UI is genuinely approachable, with a wizard that walks beginners through camera setup, while professionals get the full surface: multiple audio/video inputs (USB, RTSP network cameras, virtual devices), codec and processing settings, separate audio muxing, and hardware acceleration via Nvidia CUDA, Intel VAAPI, or Raspberry Pi. Serving your own audience is first-class - a built-in Video.js player embeds in your site, a ready-made publication website streams without any embedding work, HLS chunk sizes are tunable, and automatic Let's Encrypt handles HTTPS. Viewer and bandwidth monitoring with limits keeps traffic costs predictable, and it's GDPR-friendly: no third-party provider, no audience data stored. A fully Swagger-documented REST API drives automation. SRT support keeps latency under a second.

AzuraCast is a solution for running internet radio stations, packaging the entire broadcast stack into a single Docker installation that gets you on air in minutes. Built on PHP with a Vue.js frontend, it uses Liquidsoap as the AutoDJ engine to compile media from playlists, live DJ inputs, and scheduled programming into composited output streams, then broadcasts via Icecast-KH, SHOUTcast 2 DNAS, or Rocket Streaming Audio Server to thousands of concurrent listeners. The media library supports drag-and-drop uploads with automatic metadata extraction, while playlist management handles sequential, random, weighted, and time-scheduled rotations. Live DJ accounts let remote broadcasters connect via BUTT, Mixxx, or any Icecast-compatible software, with automatic crossfade transitions between AutoDJ and live input. The built-in Web DJ tool enables broadcasting directly from the browser without additional software. Each station includes public-facing player pages with now-playing metadata, album art, song history, and embeddable widgets for external websites. The listener request system lets audiences queue songs through the public interface. Multi-station administration hosts unlimited stations on a single server with individual user accounts and granular role-based permissions. Comprehensive analytics track listener counts, geographic distribution, unique listeners, and listening duration with exportable reports. HLS streaming delivers adaptive bitrate broadcasts for mobile compatibility. LADSPA audio plugins enable professional equalization and processing in the stream pipeline. 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.

Lightpanda is a headless browser built for AI agents and automation that loads and runs web pages without ever drawing them, so crawling, scraping, and agent browsing finish far faster and lighter than in Chrome. You drive it with the tooling you already have: Puppeteer, Playwright, and Chromedp connect over the Chrome DevTools Protocol, Selenium works over WebDriver BiDi, and existing Chrome automation scripts keep running against the new endpoint. Scripts can execute JavaScript on real pages, wait for network activity, read and mutate the DOM, submit forms, and pull structured data out of dynamic sites that plain HTTP clients cannot see. An MCP server mode hands the browser to AI applications as a callable tool, so a model can open pages, follow links, and read content in the middle of an agent workflow. The built-in agent accepts plain-language browsing tasks and can export what it did as PandaScript, a replayable JavaScript file that runs later with no model call at runtime. A fetch command loads one or more URLs and prints the result as HTML or Markdown, which turns page extraction into a single shell step inside data pipelines. In the project's published crawl benchmark it processed pages roughly nine times faster than headless Chrome at a fraction of the memory, so one server can hold many more concurrent browsing sessions. 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.

With 11,000+ GitHub stars and over 1,291 releases since 2016, Misskey is the most feature-rich open-source microblogging platform in the Fediverse — federation-first social networking with capabilities that go far beyond simple status updates. The TypeScript and Vue 3 frontend delivers a highly customizable interface with drag-and-drop widget layouts, switchable themes, per-user custom CSS, and AiScript-powered plugins that transform every instance into a unique social experience. ActivityPub federation enables seamless interaction with Mastodon, Pixelfed, PeerTube, and any ActivityPub-compatible platform across the decentralized network. Emoji reactions let users respond with custom instance emojis instead of simple likes, while Misskey Flavored Markdown adds animated text effects, sparkle decorations, and rich formatting to posts. The built-in Drive provides personal cloud file management with folder organization for images, videos, and audio — eliminating the need for external storage services. Antennas create custom filtered timelines based on keywords, users, or criteria for monitoring specific topics. Channels offer topic-specific feeds separate from the main timeline. Pages enable users to build personal websites directly within their instance. Clips save and organize posts into curated collections. The Node.js backend scales with PostgreSQL for persistence, Redis for caching, and optional S3-compatible storage for media. Docker Compose deployment provisions the complete stack including the database and cache in under 15 minutes. 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.

Self-hosting a language model is only half the job; people still need somewhere pleasant to talk to it, and Chat UI is that window, the same interface that powers HuggingChat. Point it at any OpenAI-compatible endpoint, your own Ollama or llama.cpp server, the Hugging Face router, or a hosted gateway like OpenRouter, and the models it serves appear in the picker automatically, with no per-model configuration. Conversations are saved to your database and stay browsable in the sidebar, so a thread from last week can be reopened, continued, and compared against a fresh answer from another model. Multimodal models accept image uploads right in the composer, and models that support it can call MCP tools mid-conversation to fetch data or run actions. For a shared team deployment, OpenID Connect sign-in keeps individual histories and settings separate, and per-user defaults travel with each account. Everything a consumer chatbot would keep on someone else's servers, prompts, uploaded files, and usage history, stays in the MongoDB you control on the same machine. 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.

Run a Stack Overflow of your own: Apache Answer brings the question-and-answer format in-house, maintained under the Apache Software Foundation with an Apache-2.0 license. You use it to run a community forum, product help center, or internal knowledge base where content lives as questions and answers rather than wiki pages. It ships the mechanics that make that format work: voting, accepted answers, a reputation system with privilege levels, tagging, full-text search with filters, revision history on every edit, and admin/moderator/user roles. Content is written in Markdown with real-time preview and code syntax highlighting. A plugin system covers OAuth login (Google, GitHub), S3 storage, external search backends like Algolia, and Akismet anti-spam, and a REST API exposes platform data for integration. The backend is Go, the frontend React, and it runs against SQLite, MySQL, or PostgreSQL. Self-hosting replaces per-seat tools like Stack Overflow for Teams with a flat-cost instance where you own all the content.

Graduated as an Apache Software Foundation Top-Level Project with over 15,500 GitHub stars and 100,000+ Docker Hub downloads, Apache Answer delivers the structured Q&A platform that Stack Overflow and Discourse popularized — fully self-hosted under Apache 2.0 with zero vendor lock-in. The Go backend with React frontend serves questions, answers, and knowledge articles with real-time Markdown preview using CommonMark syntax, inline @mentions to ping domain experts, and transparent revision history tracking every edit. Version 2.0 introduced AI workflows including an integrated AI assistant that helps draft and improve answers, a Model Context Protocol server for connecting AI agents to your knowledge base, API key management, and editor plugin support for extending the writing experience. Advanced search filters by tags, usernames, scores, and date ranges, while real-time suggestions surface relevant existing questions as users type to reduce duplicates. The reputation system rewards quality contributions with configurable privilege thresholds, and admin/moderator/user roles control access across the platform. A plugin architecture enables community-built extensions for third-party OAuth login, caching backends, search engines, and storage providers. Theming supports custom layouts, dark mode, and responsive design across devices, with content available in 15+ languages translated by the community. Bulk user import, email domain restrictions, and content access controls secure the platform for enterprise deployment. 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.

With over 10,000 GitHub stars and a reputation forged in the frustration of laggy corporate screen sharing, Screego delivers what Microsoft Teams and Zoom screen share consistently fail to provide: instant, high-resolution screen streaming that lets remote colleagues actually read your code. Built on WebRTC peer-to-peer connections, Screego transmits screen data directly between browsers without routing through centralized servers, achieving sub-second latency and native-resolution quality that makes pair programming and code reviews genuinely productive. The server component, written in Go with a TypeScript frontend, deploys as a single binary or Docker container with minimal configuration — set your external URL, generate a secret key, and you're live. An integrated TURN server handles NAT traversal automatically, eliminating the need for separate STUN/TURN infrastructure, with configurable UDP port ranges and support for external TURN servers when scaling beyond a single instance. Authentication supports three modes — requiring login for all actions, only for TURN connections, or none — with user management via simple credential files. Rooms are created instantly and support multiple simultaneous viewers, with system and tab audio sharing available for demonstrations that need sound. The architecture supports reverse proxy deployment behind Nginx, Caddy, or Apache with automatic TLS termination, and recent releases added riscv64 architecture support alongside existing amd64 and arm64 builds. 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.

With over 15,000 GitHub stars and positioning itself as the most feature-complete open-source alternative to Airtable, APITable combines a buttery-smooth spreadsheet interface with a full database engine and automatic API generation that turns every datasheet into a queryable REST endpoint without writing a single line of backend code. Real-time collaborative editing powered by Operational Transformation algorithms supports 100,000+ rows with multiple simultaneous users, while seven built-in view types — Grid, Kanban, Gallery, Gantt chart, Calendar, Mindmap, and Form — let teams visualize the same data in whatever format suits their workflow. The one-click API panel exposes every table as a documented REST API with filtering, sorting, and pagination, effectively functioning as a backend-as-a-service for frontend developers and automation builders. Robot automation triggers workflows on record changes with integrations into n8n, Zapier, Appsmith, and Slack, while the extensible widget system provides 20+ open-source widgets for custom charts, dashboards, and data visualizations embedded directly into datasheets. Enterprise features include SAML single sign-on, field-level and row-level permissions via Mirrors, organizational team management, audit logging, database auto-backup, and data export. Bi-directional table linking with infinite cross-links creates relational data models across workspaces, and built-in templates cover CRM, project management, inventory, and content planning. 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.

Getting a language model into the chat apps people already use takes a single self-hosted server running AstrBot, which answers in QQ, Telegram, Discord, Slack, Lark, DingTalk, LINE, and WeCom. The model itself is a setting. OpenAI-compatible endpoints, Anthropic, Google Gemini, DeepSeek, Ollama, and LM Studio all work, with fallback models and retry limits for failed requests, and Dify, Coze, or Alibaba Cloud Bailian can take over a conversation too. A bare model cannot act, so the built-in agent calls MCP servers and Skills, delegates to SubAgents, runs scheduled tasks, and executes code in an isolated sandbox. It cannot read your files either, so a knowledge base takes PDF, Word, Markdown, and EPUB uploads and searches them with hybrid dense and BM25 retrieval. Personas shape each bot's personality, and Whisper or Edge TTS adds speech. When an answer looks wrong, the ChatUI shows thinking, tool calls, token counts, and latency, and a per-message trace timeline shows how the bot got there. Because the panel steers the bot, administrator IDs, session whitelists, optional TOTP login, and HTTPS protect it. Backups live under System Settings, and the plugin market installs any of more than 1,000 extensions in one click. 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.

ILLA Builder turns internal tool development from weeks of custom frontend work into hours of drag-and-drop assembly, combining a visual canvas of 60+ UI components with direct connections to over 20 data sources. The TypeScript frontend, powered by ILLA Design, offers data grids backed by MUI X Data Grid Premium, charts, forms, rich-text editors, PDF viewers, QR code scanners, calendar widgets, map embeds, and video players. Each component supports Handlebars-style expression bindings that reference query results, widget state, or global variables, while the reactive execution tree resolves dependencies in topological order and re-evaluates downstream properties whenever upstream values change. Data connectors provide GUI-driven access to PostgreSQL, MySQL, MongoDB, MSSQL, Supabase, Redis, Firebase, CouchDB, Airtable, Google Sheets, Stripe, AWS S3, Hugging Face models, and REST or GraphQL endpoints without writing connector code. Real-time collaboration through WebSocket lets multiple developers edit the same application simultaneously with live cursor tracking and synchronized Redux state. ILLA Flow adds workflow automation triggered by schedules or webhooks, chaining actions across data sources. The Go backend handles application state, resources, and attribute-based access control using PostgreSQL for persistence, Redis for caching, MinIO for file storage, and Envoy for API ingress. Deploys as a single Docker container on port 2022 or scales horizontally on Kubernetes. 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.

Spotify economics without the subscription or catalog gaps: Navidrome, the reference self-hosted music server, streams your own FLAC, MP3, and ALAC collection from a single Go binary with a React/Material UI web player. Its Subsonic/OpenSubsonic API compatibility is the superpower: 50+ existing clients work out of the box, from Symfonium and DSub on Android to Feishin and Sonixd on desktop, plus Android Auto, CarPlay, and Android TV apps. Transcoding is server-managed and FFmpeg-backed - FLAC direct-plays at home and downsamples to MP3, AAC, or Opus over mobile bandwidth, with the OpenSubsonic transcoding extension letting clients declare capabilities and receive per-track direct-play or transcode decisions automatically. Multi-user support gives every account its own play counts, favorites, ratings, and playlists, and multi-library support scopes different collections to different users. The feature list covers serious listening: Last.fm and ListenBrainz scrobbling, artist bios and images, embedded and external lyrics, audiobook bookmarks, saved play queues that resume on another device, internet radio, jukebox mode, and M3U playlist auto-import kept in sync with your folder. Resource usage is famously low - it runs happily on a Raspberry Pi and scales to six-figure track counts.