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.

Elasticvue is a graphical interface for Elasticsearch where you can inspect cluster health, browse indices, search and edit documents, and send REST queries. Start on the Home tab, which summarizes status, nodes, shards, indices, document counts, and disk usage. Nodes shows load, CPU, RAM, heap, and disk for each member, and Shards lays out every primary and replica in a grid with a switch for shard allocation. Indices handles creation, aliases, templates, and bulk actions. Search runs query string searches across an index pattern, opens any hit in a JSON editor for updates or deletion, and downloads results as JSON. Rest sends any request from tabbed editors with history, saved queries, and examples, while Snapshots manages repositories and backups. Version 1.16 adds user and role management for Elasticsearch clusters. Secured clusters connect with basic authentication, API keys, or AWS IAM credentials, and the Docker image preloads cluster definitions from an environment variable or a mounted JSON file. Elasticvue supports every Elasticsearch version, even end-of-life releases, and speaks English, Chinese, French, Russian, Japanese, and Italian. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Mixpanel, OneSignal, and Crashlytics in one self-hosted stack - Countly is an all-in-one product analytics and engagement platform where every byte of first-party data stays on your server. A Node.js application over MongoDB, it collects through ten battle-tested SDKs spanning iOS, Android, web JavaScript, React Native, Flutter, Unity, and desktop (plus a data write API for anything else), and has powered tens of thousands of mobile, web, and desktop apps since 2012. The analytics core covers sessions, custom events, views, user profiles, and real-time dashboards, with exploration tools built for product managers as much as analysts. What separates Countly from pure analytics tools is acting on the data without third parties: built-in push notifications send automated, transactional, and personalized messages to iOS (APNs), Android (Firebase), and Huawei devices, with the SDK handling token retrieval and permission flows automatically; crash reporting captures symbolicated native crashes on iOS and Android plus JavaScript errors, correlated with the same user and session data. Email reports keep stakeholders updated, and the plugin-based architecture means features load as modules. For GDPR-sensitive products, engagement without piping user data to advertising companies is the entire point. AGPL-licensed server, installable in minutes.

Achieving 98.4% Recall@5 on LongMemEval-S without a reranker, AKB is the agent-native knowledge infrastructure giving Claude Code, Cursor, Windsurf, and any MCP-aware agent persistent organizational memory — a drop-in replacement for Confluence or Notion where AI agents read and write directly rather than scraping human-formatted pages. Documents are stored as Markdown with YAML frontmatter in Git bare repositories (full version history, branching, diff), while PostgreSQL 16 serves as source of truth for chunk text, metadata, and BM25 vocabulary, with dense embeddings handled by a pluggable vector store (pgvector default, Qdrant optional, or managed Seahorse). Agents interact through 20+ MCP tools — akb_put, akb_search, akb_browse, akb_relations, akb_sql, akb_edit, akb_history — performing hybrid dense plus BM25 search fused via Reciprocal Rank Fusion in one call. The URI graph (depends_on, related_to, implements relations) lets agents traverse knowledge connections without external graph databases. Multi-tenant vault isolation is enforced at the PostgreSQL ACL level with per-user roles and SET LOCAL ROLE for SQL sandboxing. Event fanout streams structured events to Redis Streams for external consumers (synthesis bots, digest agents, audit trails), and a hash-chained append-only audit log integrates with enterprise SIEMs. Ships with agent plugins for Claude Code and Codex (session capture, wiki ingest, lifecycle hooks), Kubernetes kustomize manifests, and Open Knowledge Format export. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. BSL 1.1 licensed (free under 100 seats, auto-converts to Apache 2.0 after four years).

1FlowBase gives teams a single self-hosted foundation that connects model providers, retains every AI conversation, and turns that activity into working business applications. You connect OpenAI, Anthropic, Gemini, DeepSeek, Alibaba Cloud Bailian, or any OpenAI-compatible API through provider plugins and publish one stable endpoint for clients. Requests convert between Anthropic Messages, OpenAI Chat Completions, and OpenAI Responses, so applications keep working when a provider changes. You can route traffic by model name, fall back across branches when a provider fails, and combine planning, execution, and review models into a virtual model that callers use like any other model. Every gateway conversation is recorded with its model, reasoning effort, status, token totals, recorded cost, and latency, then broken down by model and by user in statistics views. You define data models and data sources, expose CRUD APIs, schedule workflows, and publish an OpenAPI 3.1 document for each application. Agents reach the same runtime through MCP with progressive list, get, and call tool discovery, and you manage MCP instances and third-party servers from the console. Pages come from native React blocks that you edit in place with a built-in AI assistant. 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.

Alarik is a self-hosted S3-compatible object storage server that lets you store files, backups, media, and application data on your own hardware while keeping the S3 APIs your tools already expect. Create buckets, upload objects with multipart support for large files, and retrieve them again through core operations, range reads, presigned URLs, and bucket or object tagging. Versioning keeps earlier copies, delete markers, and listings under your control, while lifecycle rules expire old versions and clean up incomplete uploads automatically. The web console lets you drag and drop files or whole folders, search recursively across nested folders, edit metadata in place, preview individual versions, and share time limited public links. Administrators manage users, access keys, bucket policies, and public access blocks from the same console, with live charts for CPU, memory, traffic, and storage. Bucket webhooks send event notifications when objects change, and replication rules continuously copy data to remote S3 compatible targets, synchronously or asynchronously per rule. Clusters add erasure coded storage across nodes, read repair, bit rot scrubbing, and automatic rebalancing when nodes join, drain, or return. 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.

Paste some text, hit save, share the link - Hasty Paste is a fast, minimal pastebin written in Python on the async Quart framework, named, by its author's own admission, "because you use it so fast without a care in the world." No accounts, no authentication, no friction - built for the everyday case of handing a debug log, config snippet, or stack trace to someone in tech support or a chat channel. Pastes get randomly generated IDs, with an optional "long" ID mode that makes brute-force URL guessing impractical, and expiry times ensure throwaway content actually gets thrown away. Optional Pygments syntax highlighting makes code readable, a dark theme spares your eyes, and the whole interface works without JavaScript - it renders in terminal browsers and under the strictest script blockers. Storage is a custom flat-file system (no database), caching runs internally or through Redis for busier instances, and a REST API plus the companion "Hastily Paste It" CLI script enable piping command output straight into a paste from the terminal. The Alpine-based Docker image is tiny, resource usage is minimal, and the license is AGPL.

One CLI command deploys a complete AI development environment: n8n, Flowise, Supabase, Ollama, Qdrant, LiteLLM, Langfuse, Open WebUI, LightRAG, Browser-Use, Firecrawl, Crawl4AI, and more, all pre-wired with networking, credentials, and database connections. LLemonStack eliminates the hours of Docker Compose configuration that typically precede any local AI agent project. The llmn CLI initializes isolated project environments with auto-generated secure credentials, starts services in dependency order (databases first, then middleware, then apps), and displays a dashboard showing every service URL and access token. n8n brings 400+ workflow integrations, Flowise provides visual agent building, Ollama runs local LLMs like Llama and Mistral, Qdrant stores vectors at high performance, Open WebUI offers ChatGPT-style model interaction, and LiteLLM proxies requests to any provider with cost tracking. Langfuse automatically logs traces for every LiteLLM query, providing full observability. Each project maintains isolated Postgres schemas preventing data collision across parallel stacks. Firecrawl and Crawl4AI extract web content into LLM-ready formats for RAG pipelines feeding LightRAG or Qdrant. Dozzle streams live container logs for debugging. Import/export tooling migrates workflows between projects with automatic credential reconfiguration. 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 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.

Hosting finished documentation for many projects at once, Docat leaves every uploaded version online for the readers still on version 2. Zip the static HTML that MkDocs, Sphinx, or mdBook produced. Send it to a project and version with curl or the docatl command line client. Tag the newest build as latest, and links get a stable address. Readers land on a home page that opens on starred favourites. A search box narrows the list, and its queries can be shared as links. Instance statistics report documents, versions, and storage. Inside any docs page, a version dropdown with a latest label moves between releases. Uploads are open at first. Claim a project and it returns a token that must accompany every later version upload or deletion. The same token hides versions from the dropdown and from search, and each project can carry its own icon. Operators can mount a small config file for custom header and footer HTML, raise the 100 MB default archive limit with an environment variable, and put nginx basic authentication in front of the write endpoints. Swagger UI and ReDoc pages at /api/docs and /api/redoc document the HTTP API for pipeline authors. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Shell Online lets a terminal session running on your own computer follow you to any browser, so your coding agent keeps working while you are away from your desk. Prefix the command you already run, such as shell claude, shell codex, or shell opencode, and your terminal prints a link, a password, and a QR code that open the same live session on a phone, tablet, or another computer. You watch output as it happens and reply to prompts from wherever you are, and closing the browser leaves the process running until it exits on its own. Inviting a teammate takes the same link: share it read-only so they can watch without typing, or let them type into the session with you. Terminal traffic is end-to-end encrypted by default, file access stays off until you deliberately expose a chosen directory, and a scoped MCP grant can let another agent inspect or steer a session for a limited time. With an account, linked machines and their sessions gather in one web app with team sharing. The relay the sessions pass through can run on your own server with the standalone Docker deployment, pointed at by the unchanged CLI. 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 dashboards deserve version control, code review, and reproducible builds, just like the rest of your stack. DAC lets you define interactive data dashboards in YAML or TSX, validate them in CI, and serve them from a single Go binary that embeds a full React frontend. Choose from 21 chart types including line, bar, area, pie, scatter, bubble, funnel, sankey, heatmap, calendar, sparkline, waterfall, gauge, treemap, radar, and candlestick, plus metric cards, data tables, text blocks, and image widgets. The built-in semantic layer lets you define metrics and dimensions once in reusable model files, then reference them from any widget while DAC generates the SQL automatically. Connect to Postgres, MySQL, Snowflake, BigQuery, Redshift, Databricks, and DuckDB through standard Bruin connection configs. Interactive filters with date pickers, dropdowns, multiselects, and search inputs inject values via Jinja templating and re-execute only affected widgets. Live reload via Server-Sent Events refreshes connected browsers instantly when you save a file. Export dashboards as self-contained static HTML with baked-in query results for S3, GitHub Pages, or offline sharing, and render slide decks via the Google Slides export command. On RepoCloud, deploy DAC on a dedicated VPS with root SSH access and persistent storage for your dashboard definitions and database connections under the AGPL-3.0 license.

SpamEater gives you a disposable inbox on your own domain that burns itself down 24 hours after you create it. Type any name, or roll the dice for a random one, and mail sent to that address lands in a clean web inbox that refreshes itself every few seconds. Open a message and HTML mail renders faithfully, adapted to the dark interface, with inline images embedded and remote images blocked until you ask for them, so tracking pixels never fire behind your back. A live countdown in the corner tells you exactly how long the inbox has left, and a copy button grabs the address for the form you are about to fill. There are no accounts to create, no passwords to remember, and nothing to recover once the timer ends: the inbox, the messages, and the paper trail vanish together. Behind the scenes the service takes its own security seriously, with a web application firewall, intrusion prevention, rate limits, and mail bodies encrypted at rest. Dark and light themes switch with one toggle. Developers use it to test signup flows and password resets, and everyone else uses it to keep a real address out of reach of newsletters. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

A clean alternative to n8n and Zapier that puts AI-native workflows front and center, Fynt is the MIT-licensed workflow automation platform where you drag nodes onto a React Flow canvas, wire them together with typed connections, and let a distributed BullMQ engine execute them — with every step streamed live to your browser over WebSockets so you never wonder what happened inside a run. The node library spans triggers (manual, webhook, cron scheduling), AI providers (OpenAI, Anthropic, and Gemini with structured output parsing), service integrations (GitHub webhooks, Notion databases, Slack channels, Discord bots, arbitrary HTTP endpoints), logic gates (conditional branching, filtering, data transformation), and utility nodes (delay, logging, payload mapping). The execution architecture separates concerns across three services: a Next.js 16 web application serving the React 19 workflow builder and tRPC API, a BullMQ worker processing jobs with node-level retries and checkpointing for crash-safe execution, and a dedicated WebSocket relay server broadcasting real-time run events via Redis Pub/Sub. All stored credentials are encrypted at rest using AES-256-GCM with per-credential initialization vectors. The full-stack deploys as a single Docker Compose stack (web, worker, realtime, PostgreSQL, Redis) or individually via pnpm workspaces in a Turborepo monorepo. Authentication is handled through better-auth with configurable OAuth providers. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Rustunnel turns a service running on your laptop or private network into a public HTTPS, TCP, or UDP endpoint without opening inbound ports on the machine where it lives. Point the client at a local port and share the public address for webhook testing, a quick demo, an AI agent tool, SSH access, a game server, or an IoT device that cannot accept inbound connections. Give tunnels custom subdomains on your own wildcard domain, terminate HTTPS with automatic Let's Encrypt certificates, and keep client traffic encrypted from edge to local service. Put several backends behind one subdomain or TCP port and let TCP or HTTP health probes remove sick members before visitors hit them. When two peers can reach each other directly, start a P2P tunnel with NAT hole punching and QUIC, with relay fallback when the direct path fails. Watch sessions, tunnel health, traffic, and request logs from the live dashboard, inspect headers and bodies in the local request inspector, and replay a captured call while debugging a webhook. Manage tunnels, tokens, groups, captured requests, and history through the REST API, scrape Prometheus metrics, review audit logs, or let an MCP-compatible coding assistant create and close tunnels in plain language. 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.

Kuzzle ships a production-ready backend from day one: data storage with Elasticsearch search, real-time pub/sub notifications, user authentication, role-based permissions, and a multi-protocol API serving HTTP, WebSocket, and MQTT simultaneously on port 7512. Rather than rebuilding these standard features from scratch for each project, you get a ready-made backend that accepts connections from web applications, mobile clients, and IoT devices through one unified API surface. Elasticsearch provides persistent document storage with full-text search, geospatial queries, and aggregation pipelines, while Redis handles internal caching and memory storage operations. The Koncorde real-time engine powers pub/sub messaging and database notifications, enabling clients to subscribe to document changes with filter conditions that trigger instant updates when matching data is created, modified, or deleted. User management includes JWT-based authentication, role-based access control with granular permissions at the controller and action level, and pluggable authentication strategies via Passport.js. The extensible plugin framework allows developing custom business logic, additional authentication providers, storage connectors, and protocol handlers. Client SDKs for JavaScript, Dart, Kotlin, and C++ accelerate frontend development, while the Kourou CLI provides command-line API access and project scaffolding. The Vue.js Admin Console offers a web interface for managing indexes, collections, documents, users, profiles, and roles. Horizontal scaling through masterless cluster mode targets 99.99% uptime without single points of failure. 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.

bullstudio turns a pile of opaque Redis keys into a live control panel for your BullMQ and Bull queues, so you can see what is stuck, failing, or delayed without opening a Redis CLI. Point it at a Redis URL and it discovers every queue and prefix on its own, including Redis Cluster setups. The overview page tracks waiting, active, delayed, prioritized, completed, and failed counts next to throughput, failure rate, processing time, and queue delay charts. Open any job to read its input data, logs, return value, and stack trace, then retry, promote, or remove it in one click. Flow view draws parent and child jobs as a color-coded dependency graph, which makes it easy to spot the step where a multi-stage job stalled. Separate screens list connected workers with their concurrency and heartbeat, and show repeatable and cron schedules with their next run time. Pause, resume, drain, or clean a queue directly from the header. Basic authentication, a read-only mode, and a base path for reverse proxies keep access under control. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

A looping agent burns money quietly until someone can see where it drifted, and PandaProbe is an observability and evaluation platform for AI agents that makes that drift visible step by step. Capture complete executions as sessions, traces, and spans spanning model calls, tool calls, multi-agent handoffs, and custom logic, then follow a looping or failing run back to the decision where it drifted. Score traces and sessions with agent-specific metrics, run asynchronous and sampled evaluations so judge models do not cost more than inference, and use lightweight trajectory signals to catch stalls, bad tool use, and regressions over time. Bring agents built with LangGraph, CrewAI, the Claude Agent SDK, the OpenAI Agents SDK, and other integrations, with MCP tool calls captured automatically where supported. Organize work by user, organization, and project, send data through an SDK or HTTP API with project keys, and query filtered, paginated traces when you need the raw record. Keep sensitive workflows local when that matters: prompts, traces, and evaluation results stay inside the deployment you run, instead of leaving for a hosted console. 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.