BullFlow
Generic AI writing sounds like nobody in particular; BullFlow writes from your own company knowledge instead. It first builds a private knowledge base from your documents, images, audio, and video plus structured records for your company, brand, products, customers, and cases. Ingestion runs entirely on local open models: OCR reads scans, semantic splitting and chunking prepare text, and hybrid vector plus full-text search retrieves passages with their sources and relevance scores shown under every answer. A creation agent then works in one conversation: it searches the knowledge base, drafts articles with images embedded in the text, generates image sets or short videos, and exports finished pieces as PDF or Word. Output adapts to where it is going, long-form articles for article platforms, multi-image posts with matching copy for image feeds. A separate distribution gateway sends one finished piece to many accounts across Chinese and international social platforms, checks account health on a schedule, and returns publish results and metrics by webhook. Around all of that sits a commercial shell: multiple tenants with isolated data, roles from super admin to finance, credit billing measured per token, image, and second, subscriptions, invoices, referral commissions, and a conversion funnel in the operations backend. Scheduled rules can generate and publish on a timetable with nobody watching, and notifications report each completion or failure. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Signet
Give every AI agent you use the same memory, and a coding session stops starting from zero each time the tool changes. Signet runs as a daemon that captures transcripts, imported documents and connected sources, then builds a structured map of the people, projects, facts and relationships inside that history through a background process called dreaming. Each connection links back to its source passage, so an agent can trace a recalled fact to the exact transcript or file that produced it. Hooks, plugins and MCP integrations feed relevant memory into Claude Code, Codex, OpenCode, Gemini CLI, Kimi Code, Pi and more before the next prompt begins, and the same memory follows the work when the harness changes. Agents can also leave messages for each other through Signet, delivered at the start of the recipient's next session. The browser dashboard presents a daily brief, recent memories, a searchable memory graph, source connections and system status, plus a chat that answers questions over the memory with citations to the underlying records. Imports cover Word, PowerPoint, Excel, PDF, EPUB, CSV and web pages, with Notion databases and Discord channels syncing into the same graph. Per-agent read isolation, encrypted secrets and credential redaction keep sensitive material bounded. 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.
Single Server
Point it at your GitHub repos and every push goes live in under five seconds: one Linux box becomes the deployment platform for every app you have ever shipped, with no build queue, no per-app bills, and no pile of YAML. An interactive installer wires the box up once, creating a GitHub App for limited, revocable repo access, Cloudflare DNS records and a tunnel so visitors get TLS and CDN at the edge while your server IP stays hidden, and Tailscale access so you SSH in without exposing SSH to the public internet. From then on you just push: the daemon receives the webhook through a Tailscale Funnel, verifies its signature, fetches the commit, builds the image on the server from your Dockerfile or a generated one, and swaps traffic only after the new container passes its health check, so a failed deploy leaves the previous version serving. Each app runs in its own container with its own environment variables and persistent storage, and one CLI on the box handles the rest: add or remove apps, attach a domain in one command, roll back to an earlier commit, stream logs, check status, take backups, or run doctor to diagnose every layer from daemon to DNS. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Liwan
When you want honest traffic numbers without tracking cookies or stored IP addresses, Liwan puts a real-time analytics dashboard on your own server. The dashboard ranks pages, referrers, countries, browsers, platforms, and devices, charts visitors over time, and filters out bot traffic by default. Session metrics add bounce rate, average time on site, and entry and exit pages, while UTM fields attribute visits to campaigns. Collection is adjustable globally and per site: choose how visitors are grouped, whether to store country and city, country only, or no location, and how long events are kept. Drop rules discard internal, test, and local traffic before storage, and a pruning tool applies new settings to history after a dry-run preview. Tracking is one line of HTML, or your own server can send events with custom names and properties. Location can come from a MaxMind database or from Cloudflare, Vercel, and other proxy headers. One instance holds several projects and sites, with team accounts and single sign-on through Google Workspace, Microsoft Entra ID, or an OpenID Connect provider such as Keycloak or Dex. 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.
NetMap
NetMap answers the four questions every homelab owner asks in four different tools: what is on my network, what address did it get, is it still up, and why is the firewall shouting about it. Devices you add appear automatically on a visual canvas where you can drag them into place, draw the links between them, and group them into VLANs or separate sites, and the layout is still there when you come back. Behind the map, every device is watched continuously: ping times graphed over history, service checks on the ports you care about, a heartbeat strip that betrays intermittent dropouts, and an uptime percentage that settles arguments. Address management stops being a spreadsheet: define subnets, import the leases your router already handed out, see used and free addresses as a grid, and get warned before a duplicate assignment causes a mysterious outage. Firewall logs stream in live and each entry links to the device it concerns, ready to search by address, port, or protocol. When something new appears, a network scan discovers hosts with their names, hardware addresses, and open ports for review before they join the inventory. 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.
CliRelay
Anyone who pays for Claude Code, Codex, and Gemini CLI separately ends up juggling three endpoints, three sets of keys, and three kinds of rate-limit surprise, and CliRelay collapses all of that into one self-hosted gateway with a control panel. Clients speak the OpenAI Chat or Responses, Anthropic Messages, or Gemini protocol, and CliRelay translates each request for whichever upstream account is healthy. Channel groups limit where each key may send traffic, round-robin, fill-first, and session-sticky scheduling spread it, and cooldowns with automatic failover move requests off accounts that have run out of quota. Add upstreams from the /manage console through browser sign-in, device codes, or bulk credential import, covering Claude, Codex, Gemini, Antigravity, Vertex AI, AWS Bedrock, Grok, Qwen, Kimi, and any OpenAI-compatible service. Portal accounts own client keys with daily, period, and lifetime quotas, RPM and TPM limits, and per-model pricing. A monitor center scores health and charts live traffic, P50 to P99 latency, and time to first token, while every request lands in a searchable log with tokens, latency, and cost. Tenants, roles with resource.action permissions, content moderation profiles, an outbound proxy pool, and an audit trail make shared deployments workable, and the console speaks English, Simplified Chinese, and Russian. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Faved
Large link collections stay fast and organized in Faved, a private, self-hosted bookmark manager built for exactly that job. Its core is a nested tagging system that outgrows flat folders: place Go and Python under Programming Languages, color-code tags, add descriptions, pin frequent ones to the top of the sidebar, and optionally roll up child-tag items into parent views. Saving is frictionless - a lightweight bookmarklet works in any desktop or mobile browser without extensions, and Apple devices can send links through the native Share menu. Faved fetches titles, descriptions, and preview images automatically, keeps that metadata fresh over time, and flags duplicates as you save. Instant as-you-type search, flexible sorting, and bulk actions (retag, delete, refetch) keep collections of any size manageable, while customizable layouts - card, list, or table - plus a system-synced dark mode adapt the interface to your workflow. Migration is first-class: import from Chrome, Safari, Firefox, or Edge with folder structure preserved, or move from Pocket and Raindrop.io keeping tags and collections. The stack is deliberately light - PHP 8 with SQLite behind a React/Tailwind frontend - deploying via Docker with no external dependencies. All data stays local: no ads, no tracking, and no risk of your library vanishing with a discontinued service.
Libre-Closet
Your closet, cataloged like an inventory system instead of a pile you rummage through every morning. Photograph each garment once and record its name, category, brand, size, colors, washing details, and acquisition date, then filter the whole wardrobe by custom categories you define yourself, with type-ahead suggestions keeping entries consistent. Uploads get automatic background removal and WebP optimization, so every item sits on a clean backdrop that makes outfits easy to judge side by side. The outfit builder combines garments into saved looks, and the scheduling calendar lays those looks across the week while showing what you have already worn, so repeats and neglected favorites become obvious. A weather forecast rides along on each calendar day, and tapping a day opens the detail you need before dressing for an important date. Wardrobes can be shared through invite links with view-only or can-edit permissions, which suits couples, families, and stylists managing client closets. The interface installs to a phone home screen as a PWA and keeps working offline, with the UI available in seven languages. Run it wide open for personal use or switch on JWT accounts for a multi-user household, all from one container with SQLite on local disk. 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.
Calagopus
When hosting multiplayer game servers for communities or friends requires container isolation and low-overhead administration, Calagopus steps in as a modern game server management panel. Server operators use it to deploy and control Minecraft, Rust, ARK, Valheim, FiveM, and any other Linux containerized workload while reusing existing Pterodactyl and Pelican eggs without modification. From the live WebSocket console, administrators monitor real-time CPU, memory, and network graphs, autocomplete saved command snippets, inspect per-layer Docker image pull progress bars, and pop the terminal out into a secondary window. Managing configuration files and modpacks happens directly in the browser through a split-pane tabbed editor featuring real-time collaborative editing, side-by-side revision diffs with one-click rollback, a visual disk usage treemap, in-place archive browsing, and VS Code workspace mounting. Database workflows are equally streamlined: you can provision MySQL, PostgreSQL, MongoDB, or Redis instances, browse and edit rows or SQLite files in a built-in data explorer, and run read-only or writable SQL queries without external database clients. Automated schedules support conditional if-else branching, regex variable extraction from console output, crash-detection triggers, and eight backup storage drivers including S3, Restic, Kopia, ZFS, Btrfs, and Proxmox Backup 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.
SuggestArr
Instead of manually hunting for what to watch next after finishing a series on Plex, Jellyfin, or Emby, let SuggestArr turn your household's recent viewing history into automated media discovery and download requests. The platform analyzes recently watched movies, shows, and anime, queries TMDb or any OpenAI-compatible LLM (including Ollama, OpenRouter, and LiteLLM) for tailored recommendations with plain-language explanations, and dispatches requests straight to Seer, Jellyseerr, or Overseerr. Built-in AI Search lets users describe a mood or plot in natural language and request matching titles in a single click. Administrators can schedule recurring discovery jobs, import public watchlists from Trakt and MDBList, and enforce granular filters for minimum ratings, release years, genres, and regional streaming service availability so you never download content you already pay to stream. Smart request controls pause jobs when Seer queues are full, require optional human approval before submission, auto-decline stale requests, and clean up unwatched suggestions 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.
NodeDR Restaurant POS
OrderRestro runs an entire restaurant from one server on the premises: tills take orders, the kitchen watches tickets arrive, tables fill and clear, and the books close at night, all without a subscription, and it keeps working when the internet drops. Billing covers dine-in and takeaway with merged bills, discounts, gift cards, tips, and refunds, and it taxes under India GST, US sales tax, EU VAT, or Spanish IVA, with UPI QR payments confirmed by an audible alert. A visual floor plan follows every table while reservations and a waitlist manage the door, and guests scan a table QR code to order, add modifiers, call the waiter, or request the bill. Kitchen tickets route to the right station on the display system with preparation tracking, and the menu supports modifier groups, combos, seasonal menus, and scheduled availability. Ingredient inventory with recipe costing, purchase orders, goods receiving, and waste tracking keeps food costs honest, and accounting closes cash nightly with variance checks, profit and loss, a balance sheet, and bank reconciliation. Staff clock in and out, shifts are scheduled with clash checks, payroll comes from clocked hours, and loyalty points, memberships, coupons, and happy-hour campaigns keep regulars returning. Delivery runs on zones with fees, minimums, and driver assignment, chains switch branches and transfer stock between them, and eighteen reports export to print, CSV, Excel, or PDF. 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.
Pomodoist
Pomodoist pairs a task manager with a Pomodoro timer, so the list of what to do and the clock that gets it done live in the same app. Type a task the way you would say it and the quick-add bar sets its priority and estimates its Pomodoros as you type, filing it into the Inbox, Today, or Upcoming views. Plan the week on a calendar, drag cards across a kanban board, spread work along a timeline, or sort it into a four-quadrant priority matrix when everything feels urgent. Start a focus session and the timer tracks intervals against the task at hand, with configurable work and break lengths, while a 3D desktop companion focuses, rests, and celebrates along with you. Share a project and the discussion stays beside the tasks, with comments on shared items and the team's focus time visible next to the work itself. Reports chart weekly focus time, completed tasks, and habit streaks, and voice notes can become task checklists when typing is not an option. An MCP endpoint lets your AI tools manage the same projects after an OAuth sign-in, and optional Google Calendar and Telegram integrations meet the planner where you already are. The self-hosted instance unlocks every feature locally, with your own accounts and one-command backups of the whole database. 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.
LLM Wiki v2
Instead of burying your documents in a vector index you cannot inspect, LLM Wiki v2 files them into a Markdown wiki you can read, browse, and question. Drop text, Markdown, or PDF files anywhere on the page and each one is catalogued automatically, filed into a folder that says what it holds, and linked to the notes it relates to. Ask a question and the answer cites the exact notes it used, with clickable sources you can open, verify, and correct. New documents update the note on their subject instead of piling up as disconnected chunks, so what you teach it stays taught. A small classifier routes filing and questions through the folder tree, keeping the wiki quick and cheap as it grows, and you can browse the tree, follow see-also links, move or delete notes, and watch the tokens spent by both the model and the classifier. Connect OpenRouter, OpenAI, Gemini, a local Ollama, or any compatible endpoint, and expose the same wiki to Claude Code or Codex as long-term memory they can recall from and write to. Because every note is plain Markdown in a folder you own, the whole knowledge base can live in git beside your other files. 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.
Ech0
Ech0 is a self-hosted personal microblog that gives you a public timeline of short posts, links, and media on your own domain, with no ads, algorithms, or subscription in the way. Write in a Markdown editor with live preview, organize posts with tags, and let links and GitHub projects unfurl into rich cards while Bilibili and YouTube videos play inline. Attach images, audio, and video to any post, browse them in an inline gallery with built-in players, and manage uploads in a visual file manager. Visitors can like, share, and comment, you moderate from the panel with captcha protection, and RSS lets readers follow without an account. Sign in with passkeys, OAuth2, or OIDC, issue scoped access tokens, and run extra accounts under Owner, Admin, and regular roles. Media can sit on local disk or any S3-compatible bucket, and snapshots, scheduled backups, and portable capsules move the whole timeline to another instance or compile it into a static site. Ech0 Copilot recaps recent posts in one click and lets you chat with your own archive, while an MCP server, webhooks, and a documented API connect AI assistants and automations. Follow live logs in a web console, administer from a terminal UI or CLI, and install it as a PWA with dark mode and multiple 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.
Aizen
Aizen is a coding agent that actually lives on your server: one small binary you can message from your phone, which fixes code, runs the tests, and reports back before you are anywhere near a keyboard. Leave it running on the box and drive it from Telegram or Discord, approving risky edits with a tap while it works behind NAT with no inbound port to expose. In the terminal it is a single chat and agent loop with a live heads-up display of model, tokens, and context, reading files in parallel, editing symbolically through the language server, and running your shell under an approval layer with a hard floor that refuses catastrophic commands even on auto-approve. It does not get to claim done early: a verify gate runs your typecheck and test suite and sends it back to fix what it broke, git-backed checkpoints let you rewind a bad turn, and seven capability-scoped sub-agents can fan out across finding, planning, building, reviewing, testing, research, and recall when a job wants a team. Point it at any OpenAI-compatible endpoint, from hosted models to your own local server, extend it with MCP servers and slash-command macros, and let the sandbox keep every child process away from your API keys with the network denied by default. 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.
Zublo
Zublo rounds up every recurring payment quietly draining your accounts and puts them all in one place where they cannot surprise you anymore. Add each subscription once with its amount, billing cycle, next due date, and the card it charges, and the dashboard keeps the running total visible instead of letting fifteen small charges pretend to be no money at all. A calendar shows the month ahead as actual payment dates, so the week when the annual domain, the streaming bundle, and the cloud bill all land together stops being an ambush. Reminders arrive before renewals while cancelling is still an option, and statistics break spending down by category and trend so you can see which corner of your digital life keeps growing. Subscriptions in other currencies convert at synced rates, keeping foreign services honest in the totals. When you want a second opinion, an optional assistant can chat about your spending, flag subscriptions worth reviewing, and point at patterns, using a provider you choose, including a local model that never phones home. Scoped keys open the data to your own scripts, and two-factor sign-in guards the vault. 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.
JobNavigator
JobNavigator runs your whole job hunt as an automated pipeline, scraping career pages and job boards, scoring every posting against your résumés, and tracking each application to its outcome. Discovery fans out across company career pages through twelve ATS handlers, aggregators like LinkedIn, Indeed, and ZipRecruiter, and a Chrome extension that saves any posting you stumble on. Each new job is scored against your base résumés, lightly for a quick number or in full, with keyword coverage and requirement mapping that show exactly where you match and where you do not. Strong matches get a tailored résumé from eight PDF templates and a cover letter written from your persona, ready for your review before anything is sent. The extension then auto-fills application forms and drafts answers to free-text questions from your stored Q&A bank. Every application moves through stages and interviews in one tracker, while Gmail polling spots replies and updates the status for you, and Telegram delivers new-job alerts and a daily digest. A household can share one install with up to five accounts, each keeping its own jobs, résumés, and PIN, while scraping runs once for everyone. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Facet
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.