Every microservice has to find the others and learn its own settings, and Nacos is the discovery and configuration platform that answers both needs from a single server. Producers register instances with names, addresses, weights, and metadata, and consumers look services up or subscribe to changes, so a fresh instance receives traffic without hand-edited load balancer rules. Health checks drop dead instances from discovery results, while weight controls in the console shift traffic gradually during a rollout or drain a node before maintenance. Configuration lives apart from code: publish properties and YAML under a Data ID and Group, split environments with namespaces, and subscribed clients apply changes within seconds. History keeps every published version, so a typo is recoverable: rollback republishes an earlier value, and listener queries show which clients picked the change up. Gray releases push a new value to a chosen slice of instances first, so a risky flag reaches a fraction of the fleet before the rest. The registry covers AI building blocks too: MCP servers register their tools, prompts and skills carry versions and visibility, and agents publish discovery cards that gateways look up like any service. Spring Cloud, Dubbo, and Kubernetes workloads plug in through SDKs for Java, Go, Python, and other languages; run one standalone node for development or a Raft-backed cluster for production. 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.

Some questions are really about how things connect, and Neo4j is the graph database that stores your data as nodes and relationships so those connections become the thing you query. Model people, products, accounts, devices, or transactions as labeled nodes with properties, then draw typed, directed relationships between them and traverse multi-hop paths in a single Cypher query instead of stacking SQL joins. The built-in Neo4j Browser renders query results as interactive node diagrams, so you can expand a fraud ring, a dependency chain, or a recommendation neighborhood visually and share the picture with your team. Cypher reads like the picture it queries: MATCH patterns describe shapes in the graph, filters narrow them, and aggregations roll up counts, sums, and shortest paths. Store embeddings as node properties and build vector indexes over them, then combine similarity search with graph traversal so matches can be filtered by owner, connections, or freshness. Full-text indexes search names, descriptions, and documents, while uniqueness constraints keep identifiers honest at write time. Bulk-load millions of nodes from CSV files with the offline import tool, then keep data flowing in through official drivers for Python, JavaScript, Java, Go, and .NET applications. Fraud teams trace shared phones and cards between accounts, platform teams map microservice dependencies before a migration, and security analysts hunt lateral movement through identity graphs. 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.

Give every laptop, server, and cloud VM you own one private address that works from anywhere, and that is exactly the overlay network Nebula builds. Each machine runs a single small binary, presents a certificate signed by your own certificate authority, and joins a peer-to-peer layer-3 mesh where tunnels form on demand between hosts that need to talk. Lighthouse nodes, the piece you park on a VPS, keep track of where every peer can be reached and introduce hosts to each other, so machines find each other without any central VPN concentrator carrying their traffic. UDP hole punching and keepalive packets open direct tunnels through home routers and office NAT, and when two hard NATs refuse to cooperate, a relay host forwards the encrypted traffic until a better path appears. Every host enforces its own firewall, default deny on all overlay traffic, with rules keyed to certificate groups, so laptops in the user-endpoint group can reach a server's web port while other groups cannot. Ordinary hosts need no inbound ports at all, and the same binary runs on Linux, macOS, Windows, phones, and FreeBSD, making the mesh feel identical at home, in a datacenter, or across three clouds. Unsafe routes can even pull an existing office subnet into the overlay, giving legacy hardware that will never run Nebula an address on the private network. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Your prompts never have to leave the building: TextGen runs open large language models on your own hardware behind a chat interface and an API that speak fluent OpenAI. Load GGUF or Hugging Face weights through llama.cpp, ik_llama.cpp, Transformers, ExLlamaV3, or TensorRT-LLM, and switch models or backends from the interface without restarting the server. Chat mode gives you conversations with characters and system prompts, notebook mode behaves like a familiar completion editor, and both accept image attachments for vision models and documents for context. Models that support tool calling can search the web, fetch pages, and do math through single-file Python tools, and MCP servers plug in for larger tool catalogs. Start it with the API flag and the same model answers at chat, completions, and messages endpoints that mirror OpenAI and Anthropic, complete with streaming, logprobs, embeddings, and optional API key authentication. Portable builds for Linux, Windows, and macOS bundle every dependency, so the distance from download to a running model is an unzip and one shell command. Because everything is local, prompts, documents, and outputs stay on machines you control, and parallel request handling lets a single loaded model serve a whole team's tools at once. 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.

Point your routers at Akvorado and every NetFlow, IPFIX, and sFlow record crossing your network becomes a chart you can interrogate. The web console opens on a live summary of flows per second, exporter count, and traffic split by autonomous system, port, protocol, country, and address family. From there, the visualize page turns stored flows into stacked, line, grid, and heatmap time series, plus Sankey diagrams that show how traffic moves between networks, interfaces, and countries. Filters use a simple SQL-like syntax, so a question like which customers pulled traffic over a transit link last night becomes a query, not a support ticket to your future self. Because the outlet polls each exporter with SNMP or gNMI, flows arrive wearing real interface names, descriptions, and speeds instead of raw ifIndex numbers nobody can read. GeoIP lookups attach country and city to addresses, classification rules tag providers, peers, and connectivity types, and optional BMP sessions add AS paths and BGP communities to every flow. Retention is tiered: raw flows keep full detail for recent days while one-minute, five-minute, and hourly rollups stretch summaries across a year of history. A Grafana plugin is available too, so the same flow data can sit on the dashboards your team already watches. 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.

Your calendars and contacts do not need a cloud account, and Baikal is the small PHP server that keeps them on hardware you control while still syncing to every phone and laptop you own. The web admin panel is where you add users, create their calendars and address books, and reset passwords, no database editing or config files required. Each person signs in with their own username and password, so an iPhone, an Android phone running DAVx5, and a Thunderbird desktop can all stay in sync without sharing one master credential. Standard well-known redirects mean most clients configure themselves when you type the server address, and the combined dav.php endpoint serves calendars and contacts from a single URL. Events carry the full CalDAV feature set your clients expect (recurring appointments, reminders, attendees, and free-busy lookups), while CardDAV address books keep names, phone numbers, and photos in one place. Because storage is a plain database file or table set, backups are a copy you can make in seconds, and moving to a bigger host is a file transfer rather than a migration project. The same server happily runs for one person tracking a family calendar or for a small office sharing team schedules and a company address book. 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.

OpenList gathers every place your files live, local disks, S3 buckets, OneDrive, Google Drive, Dropbox, NAS shares, and forty-odd other services, into one browsable file list behind a single login. Mount each storage at its own path and the whole collection behaves like one tidy drive: documents open in the browser, whether PDF, Markdown, source code, or Office files, images get a gallery mode, and videos and audio play in place with subtitle and lyric support. Everyday file work happens in the same window: upload, delete, rename, move, and copy, including copying directly between two storages when you are migrating off a cloud plan. Offline download tasks let the server fetch links for you in the background, and multi-threaded acceleration squeezes speed out of sluggish single connections. Every folder can also be mounted as a real drive on your computer through the built-in WebDAV and FTP servers, with per-user permissions deciding who may read, upload, rename, or delete. Protected routes add passwords in front of sensitive shares, permalinks hand out direct download links, and granular user rules can hide files or open archive contents without extracting them. 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.

Write a post once in LightPost and it goes out to Bluesky, Mastodon, X, Instagram, Threads, Discord, LinkedIn, Telegram, and Facebook Pages on your schedule, with YouTube video uploads handled in the same workspace, from a dashboard running on your own server. The composer fans one message out to every connected account, including several accounts on the same platform, and each destination can receive text, images, or both within that network's limits. Drafts can be saved, edited, and duplicated, and scheduled times follow your browser's timezone so a week of announcements can be queued in one sitting. Images come in from public JPEG links or a saved library, with alt text and a preview before anything ships. Every delivery is tracked per account with links to the published posts, and when one network fails you retry just that destination instead of blasting the successful ones twice. The queue persists across restarts and overdue posts resume on their own, so an overnight reboot never eats a scheduled campaign. Built-in setup guides walk you through eligibility, credentials, and permissions for each platform, account connections can be verified or reconnected from the same panel, and your credentials stay encrypted on your disk with optional password protection and exportable post history. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

walgit is a self-hosted Git server that keeps every repository inside one S3 or GCS bucket, so the machine serving your code is disposable while your history stays durable. Point the binary at a bucket, start it, and pushing to a new repository name creates that repository on the spot, with no admin console or provisioning step in between. Developers clone, fetch, and push over smart HTTP with Git LFS support for large files, and big clones are delivered as static bundle files so multi-gigabyte repositories never bottleneck the server process. The built-in web interface lets you browse file trees, read files, review commit history, and inspect diffs directly in the browser. Per-repository policy files control protected branches, fast-forward-only references, and bypass lists, so release branches stay guarded without a separate compliance tool. A JSON API and the dependency-free repos.js SDK let scripts, internal dashboards, and automation agents read repository data programmatically, with long operations streaming progress as server-sent events. Webhooks deliver exactly-once reference events to CI systems, chat tools, and deployment pipelines whenever branches or tags move. Background checkpoints and compaction keep storage tidy on their own, while token or OpenID Connect sign-in decides who may read and who may push. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Before any AI model reads a single customer message, Obsei has already stripped out the names, email addresses, card numbers, and ID numbers hiding inside it. Obsei pulls in what customers are saying from App Store and Google Play reviews, Zendesk, Freshdesk, and Intercom tickets, GitHub issues, surveys, YouTube comments, Bluesky, RSS feeds, SQL databases, and even Gong call transcripts. Every record is pseudonymised on the way in and stored in an encrypted database that never phones home. Then the labelling starts: sentiment, intent, language, and your own custom questions, each answered with a confidence score by a zero-shot classifier running locally on your machine, or by an LLM you choose through your own keys. Answers the model is unsure about are escalated to a stronger model or queued for a human instead of being silently wrong. Related complaints roll up into stable themes with week-by-week trends, and anything mentioned by fewer than five people stays hidden, so individuals cannot be picked out of a chart. When a theme spikes, alerts and digests land in Slack, and individual findings can be routed straight into Jira, Linear, or GitHub issues. Your analysts browse it all in the built-in Studio, and your AI agents can ask questions over MCP and get cited, redacted evidence back, with any action they propose waiting for your approval. 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.

Every prompt your application sends is a small buying decision, and OrcaRouter Lite makes it a smart one instead of a habit. Point your app at one OpenAI-compatible endpoint on your server, and each request is shopped across the providers whose keys you hold: OpenAI, Anthropic, Google, xAI, DeepSeek, and dozens more. Send model set to auto and the router picks the cheapest model that can actually do the job, checking whether the request needs tools, vision, or JSON mode before it commits your money. Prefer a different tradeoff and you can switch the whole gateway between balanced, cheapest, fastest, and quality strategies, with quality picks driven by real benchmark scores rather than vibes. Clients that speak Anthropic's or Gemini's native protocols connect directly too, so Claude Code, the OpenAI SDK, and google-genai all land on the same routing pipeline. Streaming responses flow through untouched, and every answer reports which model really served it. The built-in dashboard shows recent requests, spend, latency percentiles, unreachable providers, and a savings tile that tells you what your traffic would have cost if you had blindly used a flagship model for everything. Your provider keys stay encrypted on your own box, your analytics never leave it, and if every local provider has a bad day, the hosted OrcaRouter can step in as a failover safety net. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Your scanner still emails PDFs, your backup scripts still send nightly reports, and none of them will ever learn OAuth 2.0: Email OAuth 2.0 Proxy learns it on their behalf. It listens on local IMAP, POP, and SMTP ports on your server, and when a client logs in the old-fashioned way with a username and password, the proxy swaps that exchange for a modern SASL XOAUTH2 token before the connection ever reaches the provider. The client never knows anything changed, so ancient mail apps, embedded devices, monitoring tools, and hand-rolled scripts keep sending and fetching mail through Outlook, Office 365, Gmail, Google Workspace, Fastmail, Yahoo, AOL, and any other provider that demands OAuth 2.0. The hop from your client to the proxy stays simple, while the proxy opens a fully encrypted TLS connection upstream, including STARTTLS negotiation for SMTP relays that require it. One proxy instance serves many accounts at once, with each mailbox mapped to its own local port and its tokens cached under a password you control. Optional plugins can even rewrite traffic in flight: strip Microsoft Defender Safe Links back to their real URLs, stop duplicate sent messages, or swallow test mail into a black hole. Setup is a single configuration file covering server ports, per-account scopes, and token endpoints, and the proxy then runs quietly as a headless service. 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.

Stop pasting your files into ad-riddled online converters and handing strangers copies of your documents: HRConvert2 runs the whole conversion workshop on your own server instead. Drop a file onto the browser page and pick an output format from a catalogue of 488, covering documents, spreadsheets, presentations, ebooks, comics, images, audio, video, subtitles, and archives. Built-in OCR reads scanned PDFs and images and hands back editable text documents you can search and reuse. It converts 3D models and CAD drawings, renders OpenSCAD projects with their multi-file assemblies resolved, and can even build bootable ISO disk images for x86 and ARM machines from ordinary uploads. Live streams are captured and re-encoded straight from .m3u8 playlists, and every uploaded file can be virus-scanned with ClamAV before or after conversion. Finished files go back out through temporary share links that expire on their own, so sending a converted video to a client takes one click. Each visitor works in an isolated scratch space, files delete themselves when they age out, and nobody needs an account, ever. When traffic spikes, the server watches its own CPU and memory, throttles conversions, and spreads storage across disks so it stays upright. Three interfaces, seven color schemes, and 26 languages let each user shape the page to taste, and the whole thing runs happily on a Raspberry Pi with no database at all. 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.

Sharry moves files in both directions: you hand someone a download link, and they hand files back to you, without either side juggling accounts on a commercial transfer service. Upload a file and you get a public URL that is hard to guess but open to anyone who has it; add a password as a second secret, set how long the share lives, and cap how many times it may be downloaded. Expired shares clean themselves up on a schedule. Uploads run over the tus protocol, so a dropped connection in the middle of a two-gigabyte video is an inconvenience, not a restart: send the same file again and it resumes from the last received chunk. Downloads support caching and range requests, which means recipients can stream a video in the browser and scrub its timeline instead of waiting for the whole file. The reverse direction is what makes Sharry unusual: every registered user can maintain alias pages behind their own unguessable URLs, where clients, students, or family members upload files straight to that user, with an email notification when something arrives. Everything is scriptable through a documented REST API, and files can live in the database, on the filesystem, or in S3-compatible storage. 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.

Overleaf locks its best collaboration tools behind a paid server license, and AyakaLeaf Pro is the open-source edition that hands them to your own server instead. You get the familiar online LaTeX editor, with several people typing in the same document at once, but compilation runs in an isolated Docker sandbox with a full TeX Live image, so a runaway document cannot take the service down and compile times are yours to set. Reviewers work the way they do in a word processor: track changes, comment in the review panel, and roll back through full project history with restore and download. Projects stay connected to the rest of your workflow through two-way sync with GitHub, GitLab, Gitea, and Dropbox, a git-bridge for clone, pull, and push, and reference managers through Zotero and Mendeley with searchable bibliography entries. A template gallery lets you publish house styles for your lab or department, the symbol palette inserts math notation in one click, and an AI chat assistant plus a compile-error assistant help inside the editor, while an MCP server lets external AI agents edit and compile too. Administrators get SSO through LDAP, SAML, or OAuth, domain-limited self-registration, and a panel for users and projects. 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.

Split Pro settles the question that ruins group dinners, who owes whom, with a private expense ledger that runs on your own server instead of an ad-supported app. Create a group for the apartment, the trip, or the team lunch rotation, then log each shared cost with its category, currency, date, and a photo of the receipt. Every expense splits your way: equally, by percentage, by shares, by exact amounts, or with manual adjustments when someone covered the tip. Rent, utilities, and subscriptions can repeat on a schedule so recurring bills enter the ledger on their own. The balances view shows exactly who is ahead and who owes, and debt simplification collapses a web of small obligations into the fewest possible payments. Negative expenses record refunds and corrections without breaking the math, and the activity feed keeps a visible history of every edit and deletion. Coming from Splitwise, you can import your friends and groups to pick up where you left off, and bank transactions from supported providers convert straight into expenses. On phones it installs as a progressive web app with push notifications whenever a new expense lands. Logins use email magic links, Google, or your own OIDC provider, and signups can be closed so the instance stays limited to your circle. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

LinguaCafe turns your own reading material, novels, news sites, ebooks, and video subtitles, into a private language tutor that lives entirely on your server. Import a text and every word starts out highlighted as new; as you read and mark words known, the highlighting thins out until whole pages sit clean. When an unknown word stops you, the built-in vocabulary search shows its dictionary form, so inflected verbs surface as their infinitive and nouns carry their gender, while optional DeepL translation covers anything the dictionaries miss. Marking a word adds it to your vocabulary, and multi-word selections become phrases that stay highlighted everywhere they appear again. Chapters track how many new, highlighted, and known words they contain, which makes judging a text's difficulty a glance instead of a guess. The review system cycles highlighted words through Leitner-style spaced repetition, and anything you collect can be exported or pushed straight into Anki through AnkiConnect for study on your phone. Japanese learners get kanji and radical breakdowns beside the reading pane. Everything stays local: your books, your words, and your progress never leave the machine. 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.

Certificate lifetimes are collapsing toward 47 days, which turns TLS from a yearly chore into a weekly one, and CertMate is the control plane that makes that schedule boring. It issues and renews certificates from Let's Encrypt, ZeroSSL, Google Trust Services, Actalis, DigiCert, SSL.com, Sectigo, or your own private CA, validating by DNS-01 through 29 DNS providers, HTTP-01, or a CSR whose private key never leaves the device. Renewals run automatically, earlier when a certificate's life is short or the CA moves the date through ARI, and deploy hooks push each renewed certificate where it is actually served, then probe the live endpoint to prove the new certificate answers. Discovery watches hosts and certificate transparency logs for certificates you did not issue, folding them into one inventory of what is served where and when it expires. Certificates rest on local disk or in Azure Key Vault, AWS Secrets Manager, HashiCorp Vault, Infisical, or S3 storage, with encrypted backups copied off-site. Its built-in CA issues client certificates for mTLS with OCSP and CRL, and can serve ACME itself, so certbot, lego, acme.sh, and cert-manager get internal certificates under EAB rules you define. Every action lands in a tamper-evident, hash-chained audit trail you can export and verify independently. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.