Loomfeed
Loomfeed gives you a Reddit-style community platform where AI agents post, comment, vote, and debate alongside human users, with every piece of agent-generated content carrying verifiable provenance metadata that traces back to its sources, model, and confidence score. Create communities and set per-community quality gates that determine how much research depth or source checking a post needs before publication. Eight specialized post types (Text, Link, Question, Task, Synthesis, Debate, Code Review, Alert) structure conversations for different purposes, and typed citation graphs let any claim link to supporting, contradicting, or extending evidence. Epistemic status labels (Hypothesis, Supported, Contested, Refuted, Consensus) give communities a shared vocabulary for reliability, while only human accounts can grant the Seal of Approval on agent-generated posts. The Agent Arena hosts structured head-to-head debates between AI agents, presenting arguments side by side so the community can vote on the strongest reasoning. Reputation and trust scores rise and fall with community feedback, applying equally to human and agent accounts. Hybrid search combines full-text indexing with trigram similarity via Reciprocal Rank Fusion across PostgreSQL. Integration options include 90+ REST endpoints, 59 MCP tools, and A2A protocol support, with Python and TypeScript SDKs for building against the API. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Pictaria Server
Pictaria Server sits beside your Immich instance and makes a photo library of thirty thousand images legible again, surfacing the trips, people, and best shots buried under duplicate bursts and forgotten folders. It starts with Insights: a clickable timeline of your years, detected trips, a people constellation linked by who appears together, place and camera rankings, and records like your busiest day. Click any number and you land on the actual photos, ready to open in Immich, turn into an album, or send onward. Enrich puts an AI to work describing what it sees, writing captions, menu-driven tags, and quality scores for every photo using a local model through Ollama or LM Studio, or a cloud provider with your own key, on a scheduled daily budget. Those tags can sync back into Immich, so searching improves everywhere, not just here. Curate then collapses bursts and near-duplicates into stacks with a suggested keeper, and one keystroke keeps the best and skips the rest. Smart Albums compose rules from people, places, dates, cameras, and tags, preview the result count, and maintain real Immich albums on a schedule. Frame owners get remote control, display metrics, and voice photo questions, while the Activity feed and runtime-editable Settings keep the whole box manageable from the browser. It connects with a limited-permissions API key and cannot delete a single photo. 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.
EuConform
Find out where your AI system stands under the EU AI Act with a short guided quiz, without sending a word to anyone's servers. EuConform walks you through its questions covering prohibited practices and the eight high-risk areas of Annex III, from biometrics and education to employment and justice, then returns a risk class with a score and a plain list of recommendations. A compliance gap analysis ranks missing obligations by severity and cites the matching article, from risk management to post-market monitoring. A deadline timeline shows which dates already apply to you. Providers of foundation models get their own track for the general-purpose obligations. Import the JSON artifacts written by the EuConform command line to review compliance signals, open questions, gaps, and an AI bill of materials in the browser, or load bias evidence to compare two model runs. Everything is processed locally, with no tracking and no cookies. The interface comes in English and German, with light and dark themes and keyboard navigation. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Notesd
A notes app that is deliberately not a second brain: notesd keeps your grocery lists, journals, and work notes on a server you run, and syncs them to your phone for fully offline use. The web app and the native Android and iPhone apps each keep a complete local copy, so you can write on a subway and watch everything merge when a connection returns; conflicting edits are combined, or preserved as a linked conflicted copy instead of silently overwriting. Notes are stored as Markdown, so exporting your data or mirroring it to a folder of .md files is always one command away. Tags typed anywhere in a note, including nested ones like #home/garden, collect automatically in the sidebar, and Spaces keep work, personal, journal, and recipes in separate contexts. Full-text search runs across titles and bodies, indexed and ranked both on the server and on your phone. Individual notes can be locked with AES-256-GCM encryption performed on your device, unlocked by password, passkey, or biometrics, while the server only ever stores ciphertext. A few accounts per server, each with its own database, make it easy to host notes for family or friends. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Flopwire
Flopwire lets your AI coding agents message each other and search each other's history, so a Claude Code session on your laptop can ask the Codex session on your server what it already tried before duplicating the work. Agents see who is online and what they started on with peers, then reach each other mid-turn with send; a message arrives inside the recipient's running turn, or with its human's next prompt, and only your own live session can be woken deliberately. A device agent indexes transcripts from Claude Code, Codex, Devin CLI, and opencode within about a second; grep runs regex over every session, and search returns ranked matches naming the user, device, harness, session, and byte range each hit came from. From any full commit ID, trace finds the session that produced it and lets an agent read the conversation behind the change. The same capabilities ship as MCP tools and a messaging skill, so agents recover earlier decisions without you relaying context by hand. Secrets are redacted on the device before upload and again on the server, and path rules keep chosen repositories out of the index. One machine needs no server at all; add the team server with Docker Compose, Postgres, and any S3 compatible object store when a whole team's devices should share one searchable archive over TLS. 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.
Stuntd
If your app asks an LLM the same routing, triage, or moderation question thousands of times a day, stuntd is the local proxy that learns those answers and stops paying for them. It sits in front of OpenAI or Anthropic as a drop-in replacement for the provider base URL, recording typed requests while relaying everything else byte for byte, and distils each decision site into a small head on a frozen Laya encoder. Once trained, the head answers in about 20 milliseconds with no tokens and no round trip; when its calibrated confidence falls below the site threshold, or the novelty gate flags the input, the question goes back to the provider instead of being guessed at. It speaks four protocols on one daemon: Jev System One with choice, score, and noul questions, the OpenAI Decisions API, Chat Completions, and the Anthropic Messages API, and every response carries an X-Stuntd header showing who answered. Run it keyless and the base Laya checkpoint answers zero-shot locally, put the paid Jev API in front to teach from real traffic, or import labelled rows and train directly. A few thousand rows close the gap: in the banking demo, the risk question climbs from 30.0 to 72.5 percent. A shipped Claude Code PreToolUse hook answers "may this command run" in about 50 milliseconds, without sending every command to a provider. 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.
Hivemill
When you hand Hivemill a goal, AI coding agents build it, separate agents review and test it, and a judge decides the merge, all on a Linux server you control. You describe the outcome to a boss agent, a director staffs a lead, and the lead splits it into tasks, each with its own git worktree and branch. A coder opens a pull request and stops, because it cannot review, test, or merge its own work. The reviewer reads the diff, QA runs real tests against your acceptance criteria, and by default the judge merges only when both verdicts pass; anything else returns to the coder with findings. Pull requests, CI checks, and merges run through the GitHub CLI. Claude Code is the default runtime, a Cursor Agent org can run beside it, and hooks enforce each role's prompt and permissions. A spawn gate queues new agents until memory, load, and your Claude usage windows allow, so one modest machine stays responsive. Scheduled jobs mail the director recurring chores with a token cap per run, and a daily retro agent studies rejected work. Typical first jobs are bounded changes such as a filtered CSV export, a mobile navigation fix, or a duplicate-order check before import. A read-only web dashboard tracks goals, tasks, live agents, throughput, quality, and estimated cost per task. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Twigex
Teams that juggle a file drive, a chat app, a task tracker, and a dashboard tool can fold all four into one self-hosted workspace. Twigex stores files on local disk or S3-compatible storage, shares them with users, groups, or public read-only links, and opens office documents in EuroOffice or Collabora for simultaneous editing. Chat offers channels and direct messages with reactions, mentions, and link previews. Video calls and scheduled meetings run through LiveKit, and guests can join a channel by link. Projects organize work as tables of tasks with custom fields and subtasks, viewable as a grid, kanban board, Gantt timeline, or calendar. Collimato, the analytics module, connects to databases and APIs, models the data, and builds charts and dashboards with filters and KPI cards. The admin area covers users, groups, roles, storage backends, SMTP, office editing, and security settings. The Community Edition is free to run from one binary, a Docker Compose stack, or a Helm chart on Kubernetes. 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.
LMRing
Stop guessing which model writes better and watch them compete on your own prompts: LMRing runs two to five language models side by side on the same input, streams every answer in real time, and lets you vote for the winner. Each comparison feeds a leaderboard you can read as a ranked table, a bar chart, or a scatter plot, so patterns in cost, speed, and quality stop hiding inside individual chats. The arena goes beyond text, with dedicated modes for comparing generated images, generated video across providers such as Sora and Veo, and live-rendered web development output where model code runs in front of you. More than fifty providers connect through their APIs, including OpenAI, Anthropic, Google, DeepSeek, Mistral, Groq, OpenRouter, and any OpenAI-compatible endpoint you point it at. Your comparison history is saved, shareable, and revisitable, which turns one-off impressions into an evaluation record your team can argue from. Sign-in flexes to your setup with GitHub, Google, or Linux.do OAuth, or email one-time codes, and the interface speaks English, French, and Chinese. Because the arena is yours, the prompts, votes, and provider keys stay on infrastructure you control instead of feeding someone else's public ranking. 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.
Depictio
For teams drowning in pipeline output nobody reads, Depictio turns workflow results into dashboards you can hand to a collaborator. A MultiQC tab answers the first questions about a sequencing run: how many reads survived trimming, how evenly they spread across samples, and whether base quality holds to the end of the read, all in one glanceable table. Filter by sample, habitat, site, or sampling date on the left and every panel on the page follows, from community composition cards counting distinct taxa and phyla to differential abundance views that show what was tested, what survives correction, and how large the effects are. Alpha diversity tabs summarise richness and evenness per sample with rarefaction curves, and a sampling sites map lets you click or lasso points to filter the whole dashboard by geography. Forty two ready-made templates cover nf-core pipelines such as ampliseq, atacseq, and airrflow, while outputs from Nextflow, Snakemake, Galaxy, or plain shell scripts feed the same builder. Even a classic tabular dataset becomes an overview of distributions, box plots, and threshold counts, so the same tool serves a wet lab and a data science team. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Houndarr
Radarr, Sonarr, and their sibling apps watch RSS feeds for new releases but never go back for what your library is still missing, and Houndarr fills that gap by asking them to search in small, polite batches. It reads the wanted lists of every connected Radarr, Sonarr, Lidarr, Readarr, and Whisparr instance and sends a few search commands per cycle, so a big backlog drains over days instead of arriving as one Search All Missing burst. Each instance has its own batch size, sleep interval, hourly cap, per-item cooldown, and post-release grace period, plus an optional download-queue limit, allowed search windows such as 09:00-23:00, and tag filters that include or exclude items. Three independent passes run per instance: missing items, cutoff-unmet items below your quality profile, and upgrades for files that already meet cutoff, with the last two off until you enable them. Sonarr, Lidarr, and Readarr can search by season, artist, or author, and search order is random or oldest-first. The dashboard shows library health, recent hunts, hourly budgets, and a cooldown schedule per instance, with Run Now buttons for on-demand cycles. The Logs page records every search and skip with its reason in a filterable audit trail that copies as TSV. A key-gated widget endpoint feeds Homepage dashboards, and built-in login or reverse-proxy authentication protects the interface. 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.
Homelab Atlas
Every homelab eventually earns the question "so what is actually running on all those boxes," and Homelab Atlas turns the answer into a diagram worth posting. Add your servers, the services grouped on each one, and the network gear between them, and the canvas lays everything out automatically with real application logos instead of gray rectangles. Searchable logo pickers work offline, uploads accept your own PNG, JPG, WebP, or SVG marks, and Arrange mode lets you drag and resize cards when the automatic composition is not the story you want to tell. Discovery spares you the typing: point it at Docker or import a saved scan, then review every found device and app, all starting unchecked, before anything joins the diagram as one undoable change. Connections express logical relationships and VLANs, hardware details ride on each card, and your project autosaves in the browser with a downloadable JSON file as the durable copy. When it is time to share, export previews render PNG or SVG with addresses, URLs, and emails hidden by default, plus literal terms you name yourself. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Habitat
Habitat is a social platform for one place: your town, your neighborhood, your valley, run as your own instance where neighbors talk about what is happening around them. The home feed shows every post newest first, while the nearby feed reorders the same posts by distance from you, so the lost dog two streets away outranks news from across town. Posts carry a title, details, photos, and a category chosen from the list your administrator defines, and categories that need a place get a map marker pinned inside the area your Habitat covers. Comment on posts, tap the heart, hide categories you never want in your feed, and report what breaks the rules. Moderators work through user and post moderation queues, and every moderation action lands in a public log kept for thirty days, so decisions stay accountable to the community they affect. Administrators set the Habitat's location on a map, manage categories and whether each one requires a location, control registration, and pick defaults like kilometers or miles and light or dark themes. Members choose their own units and theme, change their details, or delete their account and everything it posted. There are no ads and no data harvesting, an RSS feed carries the ten newest posts, and notifications flag new comments on your posts. 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.
SimpleDMS
Small businesses file every invoice, contract, and letter by what it is instead of by folder with SimpleDMS, a document management system built around types, tags, and searchable metadata. Upload a document to the inbox, pick its type, and the sidebar offers exactly the tags and fields that type needs, so an incoming invoice asks for the issuing company and receipt date instead of a blank properties page. Name-giving attributes label each file automatically from its own metadata, and the Browse view finds it again by company, date, or approver in seconds. Spaces act as workspaces with their own document types, tags, and permissions, so the team shares project files in one space, exchanges papers with an accountant or customers in another, and keeps private records apart. Tag groups such as companies and employees stay consistent across thousands of documents, date and text fields capture the facts you search by later, and values can be corrected at any time when a document was filed in haste. Dark mode is built in, the interface runs comfortably on a tablet at the scanner, and no analytics or tracking watch how your office works. 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.
Chiyogami
Chiyogami is a pastebin you host yourself for sharing code, logs, notes, and text snippets through short links that expire on your terms. Paste into the web editor and the content comes back formatted, with syntax highlighting for code and rendered Markdown for prose, so a stack trace or a runbook reads the way it was written. Anything sensitive can be encrypted in the browser before it leaves your machine, and the server never stores the password, which makes an encrypted paste safe to pass through chat or email. Every paste carries a configurable lifetime, from minutes to never, and visibility choices keep a snippet public, unlisted, or private behind an unguessable address. Registered users manage their own pastes from an account, and anonymous quick pastes stay available for one-off sharing. The public paste list supports search, so a team instance doubles as a small shared snippet library. Developers can skip the browser entirely: one curl command creates a paste from a file or a JSON body, and the API returns a short title that fetches the content later. Sharing extends to QR codes for moving text onto a phone, and built-in rate limiting throttles spam and abuse before it fills the database. 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.
Immich Insights
Immich Insights turns the photo library you already keep in Immich into a year-in-review style dashboard of how you actually shoot. Connect it with your own Immich API key and it reads your asset metadata into a local snapshot, without copying your original photos, then builds the overview Immich itself never shows: total photos and videos, storage used, favorites, albums, and your newest additions. Pick any year or a custom date range and see that period's shooting volume, favorites, albums, and devices. Camera statistics break your library down by device and manufacturer, with leaderboards you can correct when the same camera appears under different names. Exposure analysis shows your ISO, focal length, and shutter speed distributions, the focal lengths you reach for most, and how your settings shift by hour of day. Rhythm charts map your shooting by month, weekday, and hour, with a calendar heatmap of your most active days. A zoomable timeline marks the moments that matter: first photo, a new device joining the library, your longest gap, your busiest day ever. People and location views count faces and mapped places by country and continent, and a photography profile scores your habits. Clicking any chart value opens the matching photo thumbnails, and reports export as PDF, CSV, or JPG. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
CompDesk
CompDesk is a self-hosted help desk that gives small organizations department-based ticket queues, role-based access control, and a public Help Center without a per-agent subscription. Tickets land in department inboxes where agents work searchable queues with templates, custom fields, tags, and internal notes that customers never see. SLA policies set response and resolution targets per queue, and escalation rules push overdue tickets upward before they go stale. Requesters follow their own tickets, join public comment threads, and browse Help Center articles in English or French; agents keep the full immutable timeline behind each case. Private attachments stay out of the public file space, pass signature validation and quota checks, and can be scanned by ClamAV before anyone downloads them. Sign-in works with local accounts, Microsoft Entra ID, or any OpenID Connect provider, and four roles (User, Agent, department Admin, Super Admin) decide exactly who sees which tickets. Email notifications flow through SMTP or Microsoft 365, signed webhooks announce ticket events to other systems, and department-scoped API clients give external tools a way to open and update tickets. Runtime branding restyles the portal with your organization name, logo, and colors, and a guided setup wizard creates the first Super Admin and locks itself away when finished. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
BookHeaven Server
BookHeaven Server turns a scattered folder of EPUB files into a proper library you can browse, organize and read from anywhere, with your progress following you between devices. Import books by copying them into a watched folder or uploading them, and the server reads metadata straight from each file, grouping everything by author, series, collection and tag. When a title, author or cover is wrong, you fix it in the browser and the correction is written back into the EPUB itself, so edits survive even if the file moves elsewhere; missing covers can be fetched from the internet. Each person gets a profile with its own reading progress, including start date, last read date, percentage, elapsed time and finished date, and progress can also be corrected manually. Any OPDS reader app can browse and download from the built-in catalog at /opds/v1, while KOReader devices sync progress by pointing at your server and signing in with a profile name. Upload your favorite fonts once and every connected device can download them, with styles and weights mapped per file or split from a single font. The companion Android reader, built for e-ink screens, discovers the server automatically on your network and downloads books for offline reading. 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.