Open Source App Cloud Marketplace

Governed by the non-profit Codeberg e.V. with over 150 contributors and quarterly major releases reaching v16.0 in July 2026, Forgejo delivers a complete code collaboration platform in a single Go binary that runs comfortably on 512 MB of RAM — providing full Git hosting with SSH and HTTPS, pull requests with inline code review, issue tracking with labels and milestones, kanban project boards, wikis, and an integrated package registry supporting Docker, npm, PyPI, Maven, NuGet, and Cargo. Forgejo Actions provides GitHub Actions-compatible CI/CD with reusable workflows, OpenID Connect for secure third-party access, ephemeral runners for autoscaling, and dynamic matrix support for complex build pipelines. The platform implements ActivityPub-based forge federation allowing issues, pull requests, and stars to work across independent instances — enabling a decentralized development ecosystem similar to how Mastodon federates social networks. Organizations and team permissions provide fine-grained access control, while LDAP, OAuth2, and OpenID authentication integrate with existing identity infrastructure. Code search indexes repositories for fast discovery, LFS handles large files efficiently, and mirror synchronization imports repositories from GitHub, GitLab, and other forges. Deploy via a single static binary, Docker container, or official Helm charts with SQLite for small instances or PostgreSQL and MySQL for production workloads. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. GPLv3+ licensed.

Full double-entry accounting on your own server: Manager.io Server Edition runs the same engine as the free desktop edition as a web server, so unlimited users work in unlimited businesses concurrently, with all books on infrastructure you control. The module coverage is genuinely comprehensive - general ledger with a customizable chart of accounts, sales and purchase invoices, quotes and orders, bank and cash account reconciliation, inventory with stock tracking, fixed assets with depreciation, payroll, multi-currency with exchange gains and losses, tax codes for VAT and GST regimes, and the complete reporting stack: balance sheet, profit and loss, trial balance, aged receivables and payables, and cash flow statements. The interface is translated into more than seventy languages, reflecting a genuinely global user base. The licensing model is the refreshing part: a server license is a one-time perpetual purchase, not a subscription - twelve months of updates included, optional renewals after, no per-user tiers, no data limits, and you can downgrade to the free desktop edition anytime, so your books are never hostage. Businesses wanting collaborative accounting behind their own firewall, with data sovereignty and no monthly fees, get exactly that.

ntfy sends push notifications to your phone or desktop with a single curl command: publish a message to any topic and every subscriber receives it instantly, no signup or API key required. Over 31,000 GitHub stars and 106 releases since 2021 back a server supporting five priority levels mapped to distinct notification sounds and vibration patterns, emoji tags for visual classification, click actions that open URLs when tapped, and up to three action buttons per notification for view, HTTP callback, broadcast, or clipboard copy operations. File attachments push images from surveillance cameras, documents, or any binary payload directly to mobile devices. Subscriptions work through JSON streams, Server-Sent Events, WebSockets, or raw text, with server-side filtering by priority, tags, and message ID. Authentication enforces topic-level access control through Basic Auth, Bearer tokens, or query parameters, with a built-in user and ACL management system. UnifiedPush compatibility lets ntfy serve as a push distributor for Mastodon, Matrix, and other federated services. Web Push via VAPID keys delivers browser notifications without the mobile app. The server ships as a single statically linked Go binary or Docker image supporting amd64, armv7, and arm64 architectures, consuming 30-50 MB RAM at idle with SQLite-backed message caching. Integrations include Grafana, Prometheus Alertmanager, Uptime Kuma, Home Assistant, and Ansible Semaphore. 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 / GPLv2 dual-licensed.

A minimal Ubuntu 24.04 LTS virtual private server — no pre-installed applications, no opinionated configuration, just the operating system and root access. Start from a clean slate and install exactly what you need: web servers, databases, development toolchains, CI runners, monitoring stacks, VPN endpoints, game servers, or anything else that runs on Linux. The server ships with apt package management, systemd service supervision, and standard networking configured out of the box, giving you access to over 36,000 packages in the Universe repository alone. Connect via SSH with the key you select during deployment; a dedicated IPv4 address is included with every server at no extra charge. A browser-based serial console in the RepoCloud dashboard provides emergency access without an SSH client. Snapshots let you capture the server state before risky changes and roll back with one click. Resize CPU and RAM on demand without reinstalling. Canonical's standard security maintenance covers this LTS release through May 2029 — five years of patches for the Main repository. Ubuntu Pro extends that to ten years with additional Universe coverage, kernel livepatching, and compliance tooling, available free for up to five machines.

Share code snippets, logs, configs, and notes without registration, tracking, or ads: Lenpaste is a minimal, self-hosted, anonymous alternative to pastebin.com. It is deliberately spartan in the right ways: no accounts, no JavaScript required (the entire site works in text browsers and hardened setups), and cookies used solely to store display preferences. Pastes support syntax highlighting across a long list of languages (from ApacheConf and Arduino to mainstream stacks), configurable expiration from minutes to unlimited, one-use "burn after reading" pastes that self-delete on first view, optional author attribution, and iframe embedding for dropping pastes into other pages. The form-encoded HTTP API covers everything the UI does - create pastes with title, syntax, expiration, and line-ending normalization, fetch them by ID, and query server capabilities - making it trivial to pipe command output to your paste server from shell scripts. Server operators control maximum title and body lengths, maximum paste lifetime, rate limits for viewing and creation, search-engine indexing policy, and can lock private instances behind HTTP Basic authentication. It deploys as a single lightweight Docker container, giving your team a snippet-sharing endpoint where the content never touches a third-party service.

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

Point the gh CLI at your private Githome server and it works identically to github.com, because every REST v3 and GraphQL v4 endpoint returns byte-for-byte compatible JSON shapes, headers, Link pagination, and ETag caching. A single compiled Go binary serves the REST API under /api/v3, GraphQL under /api/graphql, git smart-HTTP transport, and a server-rendered web UI implementing the GitHub Primer design system with full functionality when JavaScript is disabled. The web interface provides repository browsing with Markdown rendering and syntax highlighting, issue tracking, pull requests with unified diffs and inline code review threads, full-text search, user profiles, and a notifications inbox. Authentication supports personal access tokens with prefix-based identification, OAuth device and web flows, and scoped permissions on every response header with only token hashes stored. The pull request system implements the complete lifecycle including async mergeable state computation, squash, merge, and rebase strategies, commit statuses, check runs, and statusCheckRollup fields. Webhooks deliver events with X-Hub-Signature-256 verification, SSRF protection, and at-least-once retry semantics. Storage supports SQLite for single-node deployments or PostgreSQL for production configurations. Performance optimizations include batch loaders replacing N+1 queries, keyset pagination, FTS-backed search, and GraphQL dataloaders. Docker deployment pulls the official multi-arch container image from GitHub Container Registry. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache-2.0 licensed.

Email discussion lists - the venerable listserv pattern where one address relays to a whole group - without the notorious setup pain of Mailman or Sympa: that is Gray Duck Mail. Its architectural trick is requiring no mail server of your own: the C#/ASP.NET Core app monitors a POP3 inbox at any external email host supporting POP3/SMTP and aliasing, relays each incoming message to all list members via SMTP, and archives it for browsing in the web interface. Replies route back to the entire group automatically. The tedious parts of list management are handled by convention: dedicated aliases process join requests, subscription confirmations, and unsubscribes, while a bounce alias in the return-path catches delivery failures so dead addresses get handled instead of silently rotting. The web administration interface creates and removes lists, manages contacts, browses message archives, and imports or exports the local database for backup. GPLv3-licensed, Docker-distributed, and fully code-documented, it targets exactly the groups email still serves best - neighborhoods, schools, congregations, workplaces, and families - where one message keeping everyone in the loop beats yet another app nobody wants to install.

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.

One download gives you a coding agent and personal assistant that is ready to work after a single message. The agent runs shell commands, reads and edits files, searches the web, calls MCP servers over stdio or HTTP with OAuth, and splits work across parallel sub-agents. Chat from the terminal, the web dashboard, a desktop app, VS Code, Obsidian, an Android app, or IM bridges for Telegram, Discord, Feishu, DingTalk, WeCom, and WeChat. Bring DeepSeek, Kimi, Qwen, Anthropic, OpenAI, or any compatible endpoint, with prompt caching tuned for each provider to keep token bills predictable. Skills use the Claude Code SKILL.md format, so an existing skills folder works unchanged. Cross-session memory, scheduled tasks, reusable multi-agent workflows, and Chrome automation that records, replays, and repairs browser steps cover the repetitive jobs. Light Apps are small agent-built tools that run in the browser with no further tokens. Safety is built in: catastrophic commands are hard-denied, deleted or overwritten files go to a recycle bin first, and an OS sandbox can confine the terminal. There is zero telemetry, and the web dashboard loads no outside fonts or scripts. Expose the server on a VPS behind its auto-generated access key and reach the same sessions from any device. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Point Claude Desktop, Cursor, Cline, or any other MCP host at one loci instance and every assistant searches the same notes, project docs, and saved memories. Hybrid retrieval fuses vector hits with BM25 using reciprocal rank fusion, so an exact identifier like user_id still surfaces, and every answer cites the file and section it came from, with optional claim-by-claim verification. Notes stay readable as Markdown, and the index updates incrementally from folders of text, HTML, org files, PDF, and Word documents or an Obsidian vault. Durable memories are saved with one command, structured task state is stored and fetched by exact key, and decision notes can declare what they supersede so a history command shows the chain and answers stop quoting retired policy. A memory health report shows which notes were cited, corrected, or never used. The web console has a dashboard, search with an evidence pane, memory editing, sources and file uploads, a query log, a note graph, and an activity log, in English or Chinese with light and dark themes. Role-based tokens separate viewers, editors, and admins. Embedding and chat calls go only to the providers you configure, and local endpoints like Ollama keep the whole workflow offline. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Long agent tasks tend to stall, lose context, or return something nobody can verify, and OpenCorvus is built to prevent all three. OpenCorvus organizes long, complex tasks into Missions worked by Expert Squads, where research, implementation, and independent review each have a named owner and hand off written artifacts. Task state and context are retained, so a stopped run can be recovered rather than restarted, and tool calls, results, and review evidence stay in a record you can inspect. Pick a squad from the bundled catalog of more than a hundred, or let the Dynamic squad compose task-specific experts for you. Connect any of roughly ninety model providers, add MCP servers and skills, schedule recurring work like a weekday brief, and receive tasks from chat channels like Slack, Telegram, Discord, and WhatsApp. A persistent project memory keeps notes the agents reuse across conversations. Start the headless server in a repository, open the workbench in your browser, or create Tasks through the HTTP API and follow their events as they stream. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Type one sentence about a feature, and Shep spins up an AI coding agent on its own branch and worktree, then carries the work through commit, push, CI, and a draft pull request. Run three features, five, or ten at once, with Claude Code, Codex, Copilot, Cursor, Gemini, Kimi Code, Cline, or local models picked per feature. A control center draws every repository and feature as a live graph with a fleet bar that counts what is cruising, queued, waiting on you, or failed. When an agent is unsure it asks a blocking question with a recommended answer and a deadline, and you reply from the web, the CLI, chat, or the terminal console. Optional spec-driven mode writes requirements, research, and a plan as YAML for you to approve before any code exists. CI watch reads failing logs and pushes fixes, a checkpoint follows every agent step so you can diff or restore it, and nothing merges without your approval. Tool pages, skills, plugins, webhooks, project memory, and per-feature chat round out the workspace, and an app starter plans a brand new project from a prompt. State stays in a local SQLite database, and usage metrics can be switched off. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Turn a rough idea into a downloadable short film without leaving one tab. OpenDirector gives you an infinite canvas where text, image, video, audio, and timeline nodes connect with reference lines, so a mood board can feed a prompt and a prompt can feed a shot. Talk to the AI assistant, type @ to pull in any asset as a reference, and watch generation tasks in a history panel for cancelling or resuming them. Image tools include smart edit, outpainting, upscaling, cutout, presets for character and product three-views, a crop and color editor, and a layer studio with masks. Open the timeline to trim, split, change speed and volume, mix audio tracks, and render an MP4 in the browser, or export JSON, FCPXML, and a media package for another editor. Connect your own provider APIs or self-hosted vLLM and vLLM-Omni services for text, image, video, and audio. Autosave keeps every canvas on disk, and the interface switches between English and Chinese. Install with Python, run opendirector serve, and open it on port 8000. 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.

Your coding agent answers a question, and instead of a wall of Markdown you get a page you can actually use. ShowAI turns replies into interactive pages built from text, tables, charts, flow diagrams, sliders, and boards, so you can drag a weight, compare options, or step through an idea and watch the result change. Every page is saved as versioned files in a local project, with undo, retained drafts, and conflict protection, and the agent reads your edits before it continues. Select any text, cell, or board object to ask the agent to expand, reorganize, or visualize it, and run several conversations on one project in parallel. When a needed component is missing, the agent builds it, and pages that work can be distilled into templates. Export a page as standalone HTML, an inline fragment, or a static site. Connect a self-hosted ShowAI Server for projects shared by administrators, editors, and viewers, with account login and device management. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Hand a coding task to an AI agent, close the browser, and come back to find it finished. Pithagoras keeps every run on the server and records each event in a log, so a tab that reconnects days later replays everything it missed and then follows the live output. Chats work in the agent's Home folder or in projects with their own AGENTS.md instructions, and the sidebar groups sessions by folder with a status dot for running, idle, error, or interrupted. Beside the conversation you get panels for browsing and editing files, reviewing Git changes and branches, running a terminal, and watching background jobs. Add model providers from a setup assistant, including local llama.cpp or Ollama servers, OpenRouter, and Anthropic, then adjust tools, skills, MCP servers, and extensions per chat. Reach the same agent from Telegram, Slack, Discord, Signal, or a webhook, with roles, approvals, and an audit log. Scheduled routines report their results to a channel, and voice mode adds local speech in both directions. The portal installs as an app, supports light and dark themes, and speaks English and German. 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.

Give every AI assistant on your team safe, structured access to your databases through one small server. DBHub exposes just two tools by default, one to run SQL and one to search schemas, which costs about 1.4k tokens of context where alternatives load nearly 20k. Point Claude Code, Claude Desktop, Cursor, VS Code, or Copilot CLI at it and they can explore tables, columns, indexes, and procedures step by step, then run queries under read-only mode, row limits, and timeouts. Define several sources in one TOML file and DBHub serves PostgreSQL, MySQL, MariaDB, SQL Server, Oracle, and SQLite side by side. Write custom tools for the queries you repeat, with typed parameters and allowed values, so models call a clean function instead of improvising SQL. Opt in to query plans and connection-pool health checks when you need them. A built-in workbench in the browser runs statements, fills in custom tool forms, and lists a trace of every recent request with timing and client. SSH tunnels and SSL reach databases behind bastion hosts, and a bearer token protects the HTTP endpoint. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

GlitchTip speaks Sentry's protocol without Sentry's operational weight - open-source error tracking that your existing SDKs already understand. The pitch is pragmatic: instrument your application with the official Sentry SDKs you already know - any language they cover - and point the DSN at your own GlitchTip instance instead. Errors, exceptions, log messages, and Content Security Policy violations flow into one place for triage, grouped into issues with stack traces, with alerts delivered by email or webhook the moment things break. Where self-hosted Sentry has ballooned into a docker-compose stack of twenty-plus containers, GlitchTip is a deliberately lean Django and PostgreSQL application a small team can actually run. Beyond errors, it bundles three more monitoring concerns: performance monitoring takes a works-out-of-the-box approach - no dashboard building, just your slowest web requests, database queries, and transactions surfaced automatically; uptime monitoring pings your sites and alerts on failures, or runs in reverse as a dead-man's-switch heartbeat for cron jobs that must check in on schedule; and log search puts application logs alongside errors for faster debugging. Unlimited projects and team members, MIT-licensed, built by Burke Software - your event volume is limited only by your own hardware.