Langflow turns LLM application development into a visual canvas where every node maps to a real LangChain primitive (chains, agents, retrievers, memory, vector stores, and tools) that teams wire together without writing integration boilerplate. The platform supports 15+ LLM provider components including OpenAI, Anthropic, Google Gemini, Mistral, Groq, Cohere, Azure OpenAI, HuggingFace, and Ollama for fully local inference, with LiteLLM proxying to over 100 additional providers through a single OpenAI-compatible endpoint. Vector database integrations cover Pinecone, Weaviate, Chroma, Qdrant, Astra DB, OpenSearch, FAISS, and Milvus, while built-in Knowledge Bases introduced in version 1.8 allow RAG pipelines without any external vector service. Multi-agent orchestration enables agent-to-agent communication with conversation management, persistent Memory Bases for cross-session context retrieval, and step-by-step reasoning visibility in the interactive Playground. Every flow automatically becomes a callable REST API endpoint via the /run route and an MCP server exposable to Claude Desktop, Cursor, or any MCP-compatible client. Every component is a real Python class that developers can customize, extend, or replace, while Extension Bundles package third-party integrations as independent pip packages for modular installation. Tool integrations include web search, Slack, Gmail, Google Drive, GitHub, and custom REST API calls. Docker deployment with PostgreSQL persistence runs on port 7860. Backed by 153,000+ stars and DataStax. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Prototype multi-agent AI systems without writing orchestration code: AutoGen Studio is Microsoft's low-code interface over the AutoGen AgentChat framework. You compose teams of LLM-powered agents in a visual Team Builder, either by drag-and-drop from a component library or by editing the declarative JSON specification directly. Each agent gets a model, a prompt, tools (Python functions), and the team gets termination conditions and an orchestration pattern, sequential or LLM-driven. The Playground runs teams interactively with live message streaming between agents, a visual control-transition graph, tool-call and code-execution tracking, and pause/stop controls, which makes it a practical debugger for agent behavior. Finished teams export as JSON for use in any Python application via the TeamManager class, or serve as an API endpoint. Any OpenAI-compatible model endpoint works, including local servers like Ollama or vLLM. Microsoft labels it a research prototype: use it for prototyping and evaluation, and build production systems on the underlying AutoGen framework.

Backed by nearly 30,000 GitHub stars and powering millions of meetings through meet.jit.si, Jitsi Meet is the open-source video conferencing platform that runs entirely in the browser with no plugins or downloads required. The WebRTC-based architecture delivers HD audio and video through the Jitsi Videobridge selective forwarding unit, while Prosody handles XMPP signaling and Jicofo manages conference focus and orechestration. End-to-end encryption uses AES-GCM 128-bit keys exchanged via the Olm protocol, protecting audio, video, and screen sharing through the Insertable Streams API in Chromium-based browsers. Moderators can split participants into breakout rooms with individual password protection, enable lobby waiting rooms for admission control, set meeting passwords, and track speaker statistics with per-participant talk time metrics. Virtual backgrounds leverage TensorFlow.js body segmentation with WebGPU and WebGL acceleration, automatically adapting quality through device tier detection. Server-side recording and YouTube live streaming run through Jibri, while Jigasi provides a SIP gateway for PSTN dial-in and speech-to-text transcription. The platform includes in-meeting chat with private conversations, polls, raise-hand functionality, emoji reactions, and tile and speaker view layouts. Mobile applications for iOS and Android complement the web client, and both web and native SDKs enable embedding video conferences into custom applications. 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.

The first container registry to reach CNCF graduated status with over 29,000 GitHub stars since VMware open-sourced it in 2016, Harbor transforms the basic Docker Distribution into a hardened enterprise registry with vulnerability scanning, supply chain signing, multi-datacenter replication, and project-level access control out of the box. Trivy scans every pushed image against the NVD, GitHub Advisory Database, and distribution-specific vulnerability feeds, with scan-on-push policies that block deployment of images exceeding configurable severity thresholds. Artifact signing through Cosign keyless signatures and Notation enforces content trust policies ensuring only cryptographically verified images reach production clusters. Policy-based replication synchronizes images and Helm charts between Harbor instances across multiple datacenters using repository, tag, and label filters with automatic retry and bandwidth throttling — enabling hybrid-cloud and disaster-recovery topologies. The RBAC model isolates projects with per-project quotas, robot accounts for CI/CD automation, webhook notifications, and audit logging that tracks every pull, push, delete, and configuration change. LDAP, Active Directory, and OIDC authentication integrate with existing identity providers, while the proxy cache transparently caches images from Docker Hub, Quay, and other upstream registries to reduce pull latency and rate-limit exposure. The RESTful API with embedded Swagger UI, tag retention policies, garbage collection scheduling, and immutable artifact rules complete the lifecycle management. Deploy via Docker Compose or 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. Apache 2.0 licensed.

Half the architecture sketches on the internet trace back to Excalidraw - the MIT-licensed virtual whiteboard whose hand-drawn aesthetic made technical diagramming feel approachable, at roughly 85,000 GitHub stars. The infinite canvas offers rectangles, ellipses, diamonds, arrows with smart binding and labels, free-draw, text, images, and an eraser, with full undo/redo, zoom, dark mode, and keyboard-first ergonomics. Community shape libraries add thousands of pre-built elements - AWS architecture icons, flowchart stencils, UI wireframe kits - and everything exports to PNG, SVG, the clipboard, or the open .excalidraw JSON format that keeps drawings diffable and portable. Live collaboration works on a share-a-link model with live cursors and a laser pointer for presenting, and it is end-to-end encrypted by design: the room key travels in the URL hash, which never reaches the server, so the WebSocket relay only ever sees ciphertext. The architecture is remarkably light - the app is a static bundle served by Nginx, drawings persist locally in the browser, and the stateless excalidraw-room relay handles multiplayer - so a self-hosted deployment gives unlimited boards and collaborators with near-zero resource cost, replacing per-editor whiteboard subscriptions.

World Monitor replaces twenty or more browser tabs by fusing geopolitical, military, financial, and infrastructure signals onto a single interactive canvas built with TypeScript, Vite, globe.gl with Three.js for the 3D globe, and deck.gl with MapLibre GL for the flat map. World Monitor ingests live data from 530 upstream sources including ACLED and UCDP for conflict events, OpenSky Network for military and civilian aircraft, AISStream for vessel positions, NASA FIRMS for satellite fire detection, USGS for earthquakes, and FRED, IMF, BIS, and Finnhub for macroeconomic and market data. The Country Instability Index v8 computes real-time stress scores across 31 Tier-1 nations, while the finance radar tracks 29 stock exchanges, commodities, and cryptocurrency with a 7-signal market composite. Six specialized dashboard variants — World, Tech, Finance, Commodity, Energy, and Happy — deploy from a single codebase. AI summarization runs locally through Ollama and LM Studio integration or optionally via Groq and OpenRouter cloud providers, with Transformers.js powering browser-side inference. The platform supports 26 languages with native-language feeds and RTL layout, and provides MCP server integration for AI agent connectivity. 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.

Real-time collaborative Markdown behind your own firewall: HedgeDoc (formerly CodiMD, descended from HackMD's open-source edition) keeps team notes on team infrastructure. Share a note's URL and collaborators are editing together instantly - live cursors, changes appearing keystroke by keystroke - in a three-mode interface that toggles between raw Markdown, rendered preview, and side-by-side split. The Markdown dialect is extended where engineers need it: Mermaid, Graphviz, and Vega-Lite diagrams, MathJax for equations, syntax-highlighted code blocks, embedded content, and a presentation mode that turns a note into reveal.js slides with a single YAML header. A dropdown permission system controls each note - freely editable, limited to signed-in users, or locked read-only - and published notes become clean read-only pages for wider distribution. Revisions track every change with the ability to revert to any earlier version. The AGPL-3.0 codebase is light enough to run on a Raspberry Pi and deploys via Docker with PostgreSQL, MySQL, or SQLite. Authentication covers LDAP, SAML, OAuth2, and email. It deliberately stays a focused document editor - no page trees or kanban - and does that one job with excellent keyboard-first ergonomics for meeting notes, RFCs, runbooks, and shared scratchpads.

Backed by nearly 80,000 GitHub stars and recognized as one of the most popular developer tools in the open-source ecosystem, Hoppscotch delivers a complete API development platform that runs instantly in any browser with zero installation. The platform supports six protocols out of the box including REST with full HTTP method coverage, GraphQL with schema introspection and multi-column documentation, WebSocket for full-duplex communication, Socket.IO, MQTT for IoT messaging, and Server-Sent Events for real-time streaming. Request collections organize your API calls into logical groups with support for sub-folders, environment variables with scope inheritance, pre-request scripts and post-request test scripts written in JavaScript, and automatic code snippet generation in over twenty languages and frameworks. Teams can collaborate in real-time with shared workspaces, role-based access controls, and cloud synchronization across devices. Migration from Postman, Insomnia, or OpenAPI specifications happens with a single click import. The self-hosted edition deploys via Docker Compose with PostgreSQL for data persistence, supporting SSO authentication through SAML and OAuth providers, SMTP configuration with OAuth2, and admin dashboard for user management. The platform also ships as a native desktop application for macOS, Windows, and Linux with offline support, plus a CLI tool for running collections in CI/CD pipelines. A built-in proxy mode bypasses CORS restrictions during development, and an interceptor enables request debugging and response modification. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Agents with tool use, deep research, a document editor, an IMAP/SMTP email client with AI triage, notes, tasks, and a CalDAV-synced calendar - Odysseus bundles all of it into one open-source, self-hosted AI workspace. It runs local models through Ollama, vLLM, or llama.cpp and cloud APIs like OpenAI and OpenRouter, with a hardware-aware Cookbook that scans your machine and recommends quantized models that fit. Persistent memory uses ChromaDB with hybrid vector-plus-keyword retrieval, web search runs through a bundled SearXNG instance, and agents can use MCP servers, files, and shell access with safety controls, plus custom skills and scheduled agent tasks. A blind Compare mode runs side-by-side model duels with identities hidden and accumulates Elo-style ratings from your votes, so model selection is based on your actual workloads rather than leaderboard claims. Deep research mode - adapted from the Tongyi DeepResearch approach - reads sources through SearXNG and produces cited reports, while the email client tags, summarizes, sets reminders, and drafts replies locally rather than through a third-party mail AI. The writing-first document editor adds AI edits, Markdown and HTML support, and version history. The stack is Python 3.11 with FastAPI, SQLite for state, and a vanilla JS frontend, licensed AGPL-3.0 with zero telemetry. Because agents can read email and execute commands, keep authentication enabled and never expose it as a public unauthenticated service.

With over 164,000 GitHub stars and one of the fastest-growing open-source projects in the AI tooling ecosystem, Firecrawl is the web context API that turns any website into clean markdown, structured JSON, or screenshots optimized for large language models. The Scrape endpoint converts a single URL into LLM-ready output with approximately 67% fewer tokens than raw HTML, handling JavaScript rendering, rotating proxies, anti-bot bypasses, and dynamic content extraction with zero configuration. The Crawl endpoint recursively scrapes entire websites from a single request with configurable depth, URL filters, and concurrent page limits. The Map endpoint discovers all URLs on a domain instantly for sitemap generation. The Search endpoint performs web searches and returns full page content from results. The Interact endpoint scrapes a page then continues working with it — clicking buttons, filling forms, and extracting dynamic content using AI prompts or code. The Agent endpoint provides autonomous web data gathering where users describe what they need in plain English. SDKs are available for Python, Node.js, Go, Rust, Ruby, PHP, Java, C#/.NET, and Elixir, with an MCP server for connecting to any AI agent or MCP client. Self-hosting deploys via Docker Compose and requires Redis and a Playwright-based browser service for JavaScript rendering. 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.

With 83,000+ GitHub stars and $18.8M in Series A funding, OpenHands delivers the leading open-source platform for AI coding agents that scored 68.4% on SWE-bench Verified with Claude Opus 4.6, outperforming Devin 2.0's publicly reported 45.8%. The Agent Canvas web UI organizes work into persistent conversations where agents edit files, run shell commands, browse the web, and execute multi-step development tasks inside isolated Docker sandbox containers. The observe-plan-act loop drives agent behavior: the Python controller manages LLM abstraction via LiteLLM routing to 100+ providers including OpenAI, Anthropic, Google, DeepSeek, Qwen, Llama, and local Ollama models. Built-in skills for code review, Docker management, PRD generation, repo-rules enforcement, release notes, and test running attach to conversations automatically via auto-discovery or trigger-based activation. The Automations system schedules recurring agent tasks with configurable templates for CI workflows, dependency updates, and documentation generation. MCP server integration enables agents to access external tools and data sources. The REST API powers an OpenAI-compatible endpoint for connecting agents to chat UIs, IDEs, and voice platforms. GitHub, GitLab, Slack, and Jira integrations enable pull request reviews, issue resolution, and team notifications. The SDK provides Python and REST APIs for embedding agents in custom tools with local or cloud execution, custom agent behaviors, and Kubernetes deployment. On RepoCloud, deploy OpenHands on a dedicated VPS with Docker socket access, persistent project storage, root SSH access, and complete control over your AI development infrastructure, all under the MIT license.

Thirteen-plus LLM providers, one unified client: NextChat (formerly ChatGPT-Next-Web) is an open-source AI chat interface built on Next.js that spans OpenAI GPT-4, Anthropic Claude, Google Gemini, DeepSeek, Groq, Azure endpoints, and self-hosted backends like Ollama, LocalAI, and RWKV-Runner. Its defining trait is minimalism - the first screen loads in about 100 KB, the desktop client is roughly 5 MB, and there is no database or user system to operate; chat history lives locally in the browser with optional WebDAV or UpStash Redis sync. The Mask system saves reusable prompt-template personas you can share and debug, long conversations auto-compress to fit context windows, and Markdown rendering covers LaTeX, Mermaid diagrams, and code highlighting with streaming responses. Plugins add web search and calculators, MCP support enables external tool calling, and Artifacts previews generated content in a separate pane. Ships as a web app, Docker image, and Tauri desktop builds for Windows, macOS, and Linux, translated into 20+ languages. MIT-licensed.

Notes on Windows, macOS, Linux, Android, iOS, and the terminal, synced through your own server: Joplin pairs its open-source clients with Joplin Server, the official self-hosted backend that replaces Dropbox, OneDrive, or Nextcloud as the synchronization target. Notes are Markdown with inline attachments (images, PDFs, audio), organized into hierarchical notebooks and sub-notebooks with cross-cutting tags, alongside to-do lists with reminders and alarms. End-to-end encryption is the headline feature: enabled in the clients, it encrypts sync payloads on-device before upload, so the server stores blobs it cannot read - genuine protection even if the host is compromised. The desktop app offers both Rich Text and Markdown editors, extended by a plugin ecosystem, custom themes, and an Extension API for writing your own scripts; a Web Clipper for Chrome and Firefox captures full pages or screenshots straight into notebooks. Joplin Server ships as a Docker image with SQLite for evaluation and PostgreSQL for production, offers a filesystem storage driver for large content, and includes multi-user support and note sharing - all free under AGPL-3.0 when self-hosted. Notes stay in an open format, so the exit path always exists.

With over 57,000 GitHub stars and deployment across thousands of organizations worldwide, Gitea is the most popular self-hosted Git service, delivering a complete software development platform in a single Go binary that runs on as little as 512MB of RAM. The platform provides full repository management with pull requests featuring inline code review, branch protection rules, merge strategies including squash, rebase, and merge commit, and automated status checks through Gitea Actions. The built-in CI/CD system uses the same YAML workflow syntax as GitHub Actions and can directly reuse over 20,000 existing Actions plugins, allowing teams to migrate from GitHub with minimal pipeline rewrites using the open-source act_runner executor. The integrated package registry supports over 20 formats including npm, PyPI, Maven, NuGet, Cargo, Helm, Composer, Conda, RubyGems, Pub, and an OCI-compatible Container registry for Docker images with no additional configuration required. Project management features include issue tracking with labels, milestones, dependencies, and time tracking, alongside kanban project boards for visual workflow management. Gitea supports LFS for large file storage, built-in wiki pages per repository, webhook integrations with Slack, Discord, Telegram, Microsoft Teams, and custom HTTP endpoints, and OAuth2 authentication with LDAP, SAML, and PAM backend support. Repository mirroring enables bidirectional synchronization with GitHub, GitLab, and Bitbucket, while the comprehensive REST and GraphQL APIs power external integrations and the official tea CLI tool. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

With over 77,000 GitHub stars accumulated in under five months since its March 2026 launch, Paperclip has become the default control plane for teams running multiple AI agents in production. Rather than juggling dozens of terminal tabs with Claude Code sessions, Codex instances, and Gemini CLI workers, Paperclip organizes all agents into a company structure with org charts, reporting lines, role-based permissions, and per-agent monthly budgets that trigger hard-stops when exceeded. The platform supports any runtime through its adapter system — Process adapters manage local CLI agents like Claude Code, Codex, Cursor, Pi, and OpenCode as child processes, while HTTP adapters trigger remote agents via webhooks to OpenClaw, serverless platforms, or custom endpoints. Heartbeat-based execution wakes agents on configurable schedules, injecting goal context, budget state, and workspace paths directly into the invocation payload. The Work and Task System provides atomic checkout with execution locks, first-class blocker dependencies, and structured work products to eliminate duplicate effort. Governance features include approval workflows, decision tracking, emergency stops, and full audit trails tracing every mutation to an actor. Deployment runs as a single Node.js process with embedded PostgreSQL locally or scales to external Postgres for production, installable in one command via npx paperclipai onboard. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.

Halo is a modern CMS that handles blogs, knowledge bases, corporate sites, and full e-commerce storefronts from a single installation, distinguished by a polyglot plugin architecture that runs both Java and JavaScript extensions on a GraalVM runtime. Built on Spring Boot 3 with a Vue.js admin console, the platform delivers a clean content management experience with Markdown and rich-text editing, live preview, scheduled publishing, draft management, media uploads, and granular multi-user role-based access controls. Plugins can be enabled or disabled at runtime without restarting the server, and the official marketplace offers over 100 free themes plus a growing library of extensions covering SEO optimization, search integration, paid content access, AI writing assistance, and Q&A chatbots. Storage backends are configurable through plugins, supporting local disk, S3-compatible object storage, and other providers. The fully documented RESTful API enables headless CMS usage for decoupled frontend architectures. Database support includes embedded H2 for quick starts, with production deployments on PostgreSQL or MySQL. Nearly 40,000 GitHub stars and over 2.5 million Docker pulls reflect the project's maturity and active development pace. A single Docker command gets a functional instance running on port 8090. 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.

Upload a report, ask a "what if" question in plain language, and MiroFish builds a small simulated society that plays out the answer before you commit to a decision. It reads your seed documents, whether news, policy drafts, financial signals, or a novel, and extracts entities and relationships into a GraphRAG knowledge graph. Entities from the graph become agents with a profession, a biography, and interest tags, and the app estimates simulation rounds from the simulated time span, with a custom mode for shorter runs. Agents post and react on two simulated platforms at once, a feed-style plaza and a topic community, while a live monitor tracks rounds, elapsed simulated time, and actions. A ReportAgent then plans an outline and writes the prediction report section by section with the InsightForge, PanoramaSearch, QuickSearch, and InterviewSubAgent tools. Afterward you can keep questioning the ReportAgent, chat with any individual agent, or send a survey question to a selected group and read their answers. Typical uses include rehearsing public relations responses, testing policy drafts against likely reactions, and exploring alternate endings for a story. The interface switches between English and Chinese, and you bring your own model: any OpenAI SDK compatible LLM endpoint works, Zep Cloud stores agent memory, and the project suggests starting under 40 rounds because runs consume many tokens. 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.

OmniRoute is an AI gateway, aggregating 338 LLM providers including OpenAI, Anthropic Claude, Google Gemini, DeepSeek, Kimi, MiniMax, and GLM into a single OpenAI-compatible endpoint at localhost:20128. The gateway catalogs over 1,200 models across 90 free-tier providers and 40 free-forever providers, automatically rotating through tier-1, tier-2, and tier-3 fallback chains when any provider exhausts its quota or returns errors. RTK plus Caveman stacked token compression reduces eligible context by 15 to 95 percent before forwarding requests, cutting API costs dramatically without degrading output quality. OmniRoute exposes its full routing engine through a built-in MCP server with 104 tools across 31 scopes over stdio, HTTP, and SSE transports, plus an A2A protocol server with six autonomous agent skills and JSON-RPC 2.0 streaming. The gateway integrates directly with Claude Code, Cursor, GitHub Copilot, Codex CLI, OpenCode, and Cline through standard base-URL configuration. Seventeen routing strategies include latency-optimized, cost-minimized, and auto-scoring modes that evaluate candidates on success rate, context fit, model fitness, quota state, and circuit-breaker health. The Next.js dashboard provides real-time provider status, usage analytics, combo chain configuration, and model catalog browsing via a responsive PWA. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.