Wrap code in a flag, then toggle it per environment, per user, per segment, or by percentage - no redeploy required: Flagsmith is an open-source feature flag and remote config platform. Every flag doubles as remote config, carrying string, number, or JSON values, so functional and visual changes ship to production without a code push. Segments are the targeting engine: build them from stored user traits, then release to beta testers first, test in production by exposing features only to internal teams, or run canary deployments to a small percentage before going wide - segment membership changes instantly without app updates. Multivariate flags split traffic across two or more variations for A/B/n testing, with flag events flowing into Amplitude, Mixpanel, or Segment for analysis. SDKs cover 15+ languages including TypeScript, .NET, Java, Python, and Go, with framework support for React, Next.js, and mobile, plus local evaluation for latency-sensitive paths. Organizations, projects, and roles keep multi-team setups tidy. Core functionality - unlimited flags, remote config, targeting, multivariate flags - is BSD-3-Clause licensed with an explicit commitment that it stays open. Self-hosting suits privacy-conscious teams: flag rules and user traits stay on your infrastructure, deployable via Docker or Kubernetes with Helm.

Describe what you want in plain language and get real SQL against your real schema: SQL Chat is an open-source, chat-based SQL client from the Bytebase team. Instead of writing queries in a traditional editor, you connect a database and describe what you want in plain language; the AI reads your schema automatically, generates SQL that references real table and column names, executes it, and returns tabular results in the conversation. Follow-up messages refine the query, so exploration becomes a dialogue - narrow a result set, add a join, change an aggregation - without retyping statements. It supports MySQL, PostgreSQL, SQL Server, TiDB Cloud, and OceanBase from one interface, and covers modification as well as reads: insert, update, and delete operations phrased conversationally. Built with Next.js and TypeScript, it deploys as a single stateless Docker container in single-user mode - connection profiles live in the browser, so there is nothing server-side to maintain. A custom AI endpoint setting routes inference through any OpenAI-compatible API, including self-hosted models, and an optional database-backed mode adds accounts and quotas for offering the tool to a team. MIT-licensed.

Document-generation code is the worst kind of code in your backlog - Carbone kills it. Its insight is separating design from data - templates are ordinary office documents (DOCX, ODT, XLSX, PPTX, HTML, even custom XML) built in LibreOffice, Microsoft Office, or Google Docs, with mustache-like markers such as {d.companyName} typed directly into the text. Send a template plus JSON from your existing APIs to the HTTP API, and Carbone returns the finished document - exported as-is or converted to PDF, XLSX, CSV, HTML, PNG, EPUB, and more via its integrated LibreOffice converter (Chromium and OnlyOffice engines are also supported for HTML-fidelity and office-format conversions). The template language goes well beyond substitution: loops over arrays render dynamic table rows, filters and aggregations run inside the document, and built-in formatters handle dates, numbers, currencies, timezones, and locales, with custom JavaScript formatters when needed. One template serves multiple languages through translation markers with auto-maintained translation files. The XML-agnostic engine means anything your document editor can design - pagination, headers, footers, nested tables, charts - survives generation intact, and Carbone guarantees no breaking changes in template syntax. Node.js-based, fast via multi-threaded LibreOffice conversion. The invoices, contracts, and reports your product owes its users become template edits, not sprints.

Upload files, get a share link, let it expire: PsiTransfer is a self-hosted WeTransfer with no accounts, no logins, and no third-party cloud with size caps and metadata harvesting. The engineering focus is large files over imperfect networks. Uploads use the tus.io resumable protocol, so a dropped connection on a multi-gigabyte video resumes exactly where it stopped once you're back online; downloads support HTTP range headers for the same resilience, and everything streams, so file size is bounded by your disk rather than memory. Files organize into upload buckets with retention you control: expire after a set time (up to weeks) or after a one-time download, with automatic cleanup when links lapse. Recipients need nothing installed - they open the link, preview files in modal views, and grab everything as a zip or tar.gz archive with one click. Buckets can be password-protected (AES-encrypted download lists), and security-through-obscurity is done properly: bucket URLs use hashed UUID tokens and stored filenames are replaced with UUIDs. An optional admin page (enabled by setting an admin password) lists bucket information and storage. The Vue.js frontend ships under 100 KB gzipped and is fully responsive. Honest caveat from the author: no end-to-end payload encryption yet. BSD-licensed, Docker-ready.

Live web search plus an LLM of your choice: Farfalle is an open-source, self-hosted answer engine in the Perplexity mold. Queries route through one of several search providers - self-hosted SearXNG for a fully independent stack, or Tavily, Serper, and Bing APIs - and the model composes a cited answer from the retrieved results. Model flexibility is the core design: run llama3, mistral, gemma, or phi3 locally through Ollama for zero per-query cost and full privacy, use cloud models like GPT-4o or Groq-hosted Llama 3 for speed, or route to any provider via LiteLLM. An Expert Search mode uses an agent that plans a multi-step search strategy and executes it for harder questions, and chat history keeps prior research sessions available. The stack is a Next.js and shadcn/ui frontend over a FastAPI backend with Redis rate limiting, shipped as a pre-built Docker image. A browser search-engine entry pointing at your instance makes it the default search from the address bar. Paired with SearXNG and Ollama, the whole pipeline runs with no external API at all.

Merge, split, compress, convert, edit, annotate, redact, OCR, and sign PDFs - BentoPDF packs over 130 tools into a privacy-first toolkit that runs entirely in the browser through WebAssembly. Files are never uploaded - processing happens in browser memory on the user's machine and disappears when the tab closes, which makes the tool GDPR-clean by architecture and safe for financial, legal, and internal documents. The engine combines WASM builds of PyMuPDF, Ghostscript, and CoherentPDF; Tesseract handles OCR with searchable text-layer output; Office conversions cover Word, Excel, and PowerPoint; and digital signatures use X.509 certificates (PFX/PEM) with the private key staying on the client. Because there is no server-side processing, deployment is a static-file exercise: a single Docker container, or any static host. A dedicated self-hosted build strips the marketing pages while keeping every tool, and air-gapped deployments are first-class - an automated script bundles the WASM modules, OCR language data, and fonts for fully offline networks. No accounts, no limits, no watermarks; TypeScript and Vite under the hood.

The drawing surface inside WebRTC conference tools like Meetzi and the LAMS online-learning platform is Whiteboard (by cracker0dks) - a lightweight Node.js collaborative sketchboard built to be embedded and customized, which also slots into Nextcloud via the External Sites app. Everyone opening the same whiteboardid URL parameter draws on the same board, with remote user cursors visible live, per-user undo/redo, and an indicator showing the smallest participating screen so nobody draws outside a colleague's view. Content handling goes beyond pen strokes: drag-and-drop or paste images and PDFs from any PC or browser, then resize, rotate, and draw over them on canvas or background; add text and sticky notes; hold Shift for angle-snapped lines and perfect squares. Every function has a keybinding - deliberately friendly to pen displays like Wacom and XP-Pen whose hardware buttons map to shortcuts. Boards save to image or JSON (with reload), export directly to Nextcloud via WebDAV, and persist across restarts with the file-database option. A REST API with bundled interactive docs allows full programmatic control, an optional access token locks down uploads, and YAML configuration tunes behavior and performance. MIT-licensed and reverse-proxy friendly.

The integrations your SaaS product offers its own users - that is what Nango, an open-source product-integrations platform, exists to build. It solves the repetitive infrastructure work behind every third-party API connection: OAuth flows, API key handling, token refresh, encrypted credential storage, rate-limit backoff, retries, and multi-tenant connection management. It ships pre-built auth configurations for 800+ APIs. Your users connect their accounts through an embeddable, white-label Connect UI, and your backend then reads or writes data through Nango's proxy, SDKs, or REST API without ever touching raw credentials. Integration logic is written as TypeScript functions covering actions, scheduled data syncs, and webhook processing - all running on one runtime with retries, checkpointing, and per-connection logs built in. Syncs pull records incrementally on a schedule, one-way or two-way, which suits RAG pipelines, search indexing, and keeping local copies of external data current. Selected actions can also be exposed as tool schemas or through a built-in MCP server, so AI agents operate on user-connected accounts without ever handling provider credentials. Auth support spans OAuth 2.0, OAuth 1.0a, API keys, basic auth, and JWT, and observability - logs, metrics, failure detection, and a reconnect flow for expired credentials - is scoped per customer connection for easier support debugging. Works with any backend language. Self-hosting on RepoCloud keeps all customer credentials and synced data on infrastructure you control, which matters for data residency and compliance requirements.

Stocks, ETFs, crypto, bonds, precious metals, and cash across every account and currency, in one privacy-first dashboard: Ghostfolio is open-source wealth management software. The deliberate design decision is no brokerage linking: positions enter by manual entry, CSV import, or the REST API, so your holdings never pass through a data aggregator. Performance is measured as return on average investment across Today, WTD, MTD, YTD, 1Y, 5Y, and Max timeframes, with benchmark comparison against indices like the S&P 500, dividend tracking, and allocation breakdowns by asset class, region, and sector. A static X-ray analysis flags concentration and other portfolio risks, and a FIRE calculator projects progress toward financial independence. Multi-currency support converts holdings using historical exchange rates, market data comes from Yahoo Finance and CoinGecko among other pluggable providers, and everything exports back out as CSV or JSON. Built with Angular and NestJS on PostgreSQL and Redis, shipped as Docker images for amd64 and ARM, with a mobile-first PWA interface, dark mode, and a distraction-free Zen mode. AGPL-licensed.

Crypto portfolio tracking that inverts the SaaS model: rotki runs on your own machine, needs no email or account for the free tier, and keeps every wallet address, balance, transaction, and tax event in a local SQLCipher database encrypted with 256-bit AES. By default nothing passes through rotki-operated servers - a design choice that matters when cloud portfolio trackers concentrate exactly the identity-linked holdings data attackers want. Centralized exchanges (Kraken, Binance, Coinbase, Bitstamp, and more) connect through read-only API keys that can see but never withdraw; blockchain accounts cover Ethereum and its L2s, Bitcoin, Solana, Polkadot, and Kusama, with ENS resolution and your choice of RPC endpoint or your own node. rotki decodes on-chain transactions into readable events across major DeFi protocols - Aave, Uniswap, Compound, Curve, Lido - and generates profit-and- loss reports for tax season with customizable accounting settings, including FIFO, LIFO, and HIFO cost-basis methods, plus CSV imports for defunct exchanges. Optional premium sync is zero-knowledge, encrypting the database on-device before upload. AGPLv3-licensed and multiplatform, with a Docker package for server deployment.

What Doodle did before ads and paywalls took over: Rallly (three L's) lets you propose a few dates, share a link, and watch an availability grid fill in - no email threads, no forced accounts, no "$6.95/month to remove ads." The availability grid makes the winning slot obvious at a glance, comments on each poll keep the "I can do Tuesday if we start late" discussion attached to the decision instead of buried in chat, and email notifications fire as votes and comments arrive. When consensus lands, finalize the winning option and everyone gets notified. The stack is modern TypeScript - Next.js, tRPC, Prisma over PostgreSQL, Tailwind - with a clean, genuinely mobile-friendly UI, dark mode, and community translations in 10+ languages. Self-hosting means unlimited polls and unlimited participants with meeting data on your server rather than a scheduling SaaS. It pairs naturally with Cal.com: Rallly answers "which time works for everyone?", Cal.com handles "book a slot on my calendar." AGPL-licensed.

Typeform's conversational format, self-hosted: HeyForm is the open-source form builder that presents one question at a time. Forms present one question at a time, which measurably improves completion rates compared to long static pages. It supports 40+ field types, from text, email, and phone inputs to picture choices, date pickers, star ratings, signatures, and file uploads. Conditional logic shows or hides questions based on earlier answers, routes respondents to different endings, and redirects to URLs, so a single form can serve multiple flows. Completed submissions land in a results dashboard with drop-off and completion analytics, and connect outward through webhooks or integrations with Zapier, Make, Google Sheets, Notion, Airtable, and Slack. Theming covers fonts, colors, backgrounds, and custom CSS, so embedded forms look native to your site rather than like a third-party widget; the JavaScript embed library renders them inline, as popups, or full-page, with shareable standalone links as the default. Team workspaces and projects with member management let multiple teams share one instance without mixing data. Self-hosting removes per-response pricing entirely - unlimited forms and submissions for flat hosting cost - and keeps lead data, feedback, and quiz answers in your own MongoDB, simplifying GDPR compliance. The stack is a NestJS server and React webapp backed by MongoDB and KeyDB, distributed under GPLv3 as a Docker image.

Subtitles are the one chore Sonarr and Radarr leave behind - Bazarr finishes the *arr media stack by automating them. It connects to both via their APIs and mirrors their libraries - it doesn't scan disk itself, it manages exactly what your *arr apps index. For every monitored episode and movie it checks existing internal and external subtitles against your language profiles, then hunts missing ones across dozens of providers - OpenSubtitles.com, Podnapisi, Addic7ed, Subscene, and many regional sources - covering 184 subtitle languages including forced/foreign-dialogue tracks. Matching is smarter than filename guessing: releases are compared by release group and source, some providers support exact file-hash matching, and every downloaded subtitle gets a percentage score. Set a minimum score per Sonarr/Radarr connection and Bazarr rejects weak matches; enable upgrades and it replaces previously downloaded subtitles when better ones surface. Out-of-sync files get fixed too - automatic subtitle synchronization realigns timing after download, triggered only below a configurable score threshold so good subs aren't touched. Per-show and per-movie language configuration, download history, manual on-demand search, and adaptive searching that throttles provider API calls round it out, all behind a clean Sonarr-style web UI written in Python. If your library serves multilingual viewers, this removes the last manual step.

High-resolution weather forecasts became a free commodity because of Open-Meteo - and this deployment puts the whole open-source engine on your own infrastructure. The public open-meteo.com service aggregates national weather models (NOAA GFS, DWD ICON, ECMWF, Meteo-France, and others) into one consistent JSON interface; self-hosting gives you that same API without rate limits, third-party dependency, or usage metering. The architecture is two cooperating services: the API server exposes forecast endpoints fully compatible with Open-Meteo query parameters - latitude, longitude, hourly and daily variables like temperature, precipitation, wind, and radiation - while a background sync worker downloads fresh weather model data on a configurable interval into a shared persistent volume at /app/data, so forecasts stay current and survive restarts without re-downloading. You control which weather models to mirror, which variables to store, how much historical depth to keep, and how often to refresh - meaning a lean deployment can sync only the model and region you actually query. Responses are plain HTTP/JSON, so integration with dashboards, Home Assistant-style automations, agricultural monitoring, IoT fleets, or any application takes minutes. For anyone making thousands of forecast calls a day, replacing a metered weather API with your own instance turns a recurring bill into a flat infrastructure cost.

Retrieval-augmented chatbots on your own knowledge base - that is the whole mission of Dialoqbase, an open-source bot-building platform. Feed it content through a broad set of data loaders - web pages and full crawls, sitemaps, PDFs, DOCX, CSV, plain text, GitHub repositories, YouTube videos, and MP3/MP4 audio - and it handles the whole RAG pipeline in one self-contained app: chunking, embedding, vector storage, and LLM querying. The distinguishing architecture choice is PostgreSQL with pgvector for embedding storage and similarity search, which removes the separate vector-database dependency, and Redis-backed Bull queues for ingesting large documents without blocking the API. Model choice is wide open: OpenAI, Anthropic Claude, Google Gemini, Cohere, Fireworks, Hugging Face, local models via Ollama, and any OpenAI-compatible endpoint, with an equally broad list of embedding providers. Finished bots embed on any website with customizable styling or deploy to Telegram, Discord, and WhatsApp, and an API creates and manages bots programmatically. Multi-user support adds registration limits and per-user bot quotas. MIT-licensed and free for commercial use.

Bring your own API keys and work with OpenAI GPT models, Anthropic Claude, Google Gemini, Mistral, DeepSeek, Grok, Azure endpoints, and local models in one organized workspace: TypingMind is a unified chat frontend for large language models, replacing a browser tab per provider. Parallel chat sends the same prompt to multiple models and compares answers side by side, and models can be switched mid-conversation. A prompt library stores reusable, tagged prompts with variables, and the AI Agents system builds specialized assistants that bundle a base model, custom instructions, assigned plugins, and uploaded knowledge files for RAG. Plugins extend every connected model with web search, image generation (DALL-E, Stable Diffusion), Deep Research, URL reading via Firecrawl, and Zapier automation - plus MCP server integrations for Notion, Atlassian, and other external tools, and a JavaScript extension API for custom behavior. Chats store locally by default with optional sync. Self-hosting puts the interface on your own domain and, for teams, adds branding, member access limits, and shared prompt and agent libraries.

Credentials sitting forever in email threads and chat scrollback - Password Pusher solves that everyday security failure. Instead of pasting a password into Slack, you push it - a password, note, file, URL, or QR code - and share a unique one-time link that expires after a set number of views, a time limit, or both. Content is encrypted at rest with AES-GCM under a configurable master key, optionally guarded by a passphrase, and permanently deleted from the database the moment it expires; a retrieval-step option keeps URL-scanning bots from consuming views. Full audit logs record when each link was created and viewed (and by whom, with logins), and TOTP two-factor authentication can be required instance-wide. The delivery page is deliberately unbranded - no logos or confusing links for recipients - and the interface ships in 31 languages with light and dark themes. Automation runs through a JSON API (v2), an official CLI for pushing and expiring secrets from the terminal, a Chrome extension, and a catalog of third-party integrations. Apache-2.0 licensed Ruby on Rails, deployable via Docker, Kubernetes, or Helm, with SQLite or PostgreSQL storage - the sysadmin staple for sending credentials that clean up after themselves.

Vehicle maintenance records shouldn't live in a homemade spreadsheet and a shoebox of receipts - LubeLogger exists because of exactly that. It's a self-hosted .NET web application (from Hargata Softworks) that gives every vehicle in your garage a complete digital history: service, repair, and upgrade records categorized and searchable, with invoices and receipts attached as documents; fuel fill-ups that compute economy in MPG, UK MPG, L/100KM, or KM/L without spreadsheet tedium; taxes, odometer logs, inspections, equipment, and supplies inventory for tracking the parts and fluids on your shelf. The reminder system is what saves engines: recurring reminders trigger by date, odometer reading, or whichever comes first - exactly how real service intervals work - so oil changes and timing belts stop relying on memory. A dashboard summarizes expenses by year, month, and category, a planner tracks to-dos by type, priority, and progress, and professional vehicle reports print a full history - genuinely useful when selling a car. Collaboration is built in: invite household members as collaborators on shared vehicles, with single sign-on support. Custom fields adapt records to your needs, CSV import/export moves data freely, an API enables automation, and the mobile-tested UI installs as a PWA on iOS and Android. MIT-licensed.