Categories
Knowledge Management Generative AI Enterprise Search AI Search Search API RAG Platform Federated Search Metasearch EngineStars
Forks
Watchers
Developer links
Swirl
Type one question and SWIRL Community queries your connected apps live, ranks what comes back, and can draft a cited answer, with no vector database and no second copy of your data. Aim it at SharePoint, Confluence, and Drive and you get one knowledge base search where every hit links back to its source. Support assistants search docs and tickets to draft grounded replies, and developer assistants answer from GitHub, Jira, and documentation. Each query runs with the user's own permissions, and Microsoft 365 connects through OAuth2 so people only see what they can already open. Results are re-scored with cosine similarity, duplicates are detected, and the Galaxy UI highlights matches above a configurable relevancy threshold. View By reorders the list by relevancy, date, or top pick, and Source checkboxes narrow it to chosen systems. Queries accept AND, OR, and NOT, and starting one with a tag such as company: searches only the sources carrying that tag. Connect OpenAI, Azure OpenAI, or an OpenAI-compatible endpoint and the Generate button writes a cited summary of the best results. Select Items picks exactly which results the AI reads, and optional instructions steer its tone, length, and focus. The admin console manages sources, AI providers, users, and API tokens, and developers call the REST API or write their own connectors and mixers. 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.
Benefits
- Answers Without Copying Your Data
- Queries run live against each source with the user's own permissions, so there is no vector database to stand up, no ETL pipeline to build, and no second copy to secure.
- Cited AI Answers From Your Sources
- Generate a summary from the top results with clickable citations through OpenAI, Azure OpenAI, or any OpenAI-compatible endpoint, and choose exactly which documents the model reads.
- First Search Works After Install
- arXiv, European PMC, and Google News come preconfigured, so a federated search returns ranked results minutes after docker compose up, before you connect a single internal system.
- Extend It With Custom Connectors
- Add SearchProviders from the admin console, write your own connectors, processors, and mixers, or call the REST API from internal tools and agents to embed SWIRL results elsewhere.
Features
- Federated Live Search
- Sends each query to every active SearchProvider in parallel, spanning databases, search engines, SaaS apps, and web APIs, then normalizes the responses into one unified result set.
- Cosine Relevancy Re-ranking
- Re-scores merged results with cosine vector similarity using spaCy and NLTK, detects duplicates, highlights matched terms, and orders by relevancy, date, or top pick.
- Galaxy Search Interface
- Web interface with a source picker, tag chips, per-source facets, Select Items for choosing RAG inputs, optional AI instructions, and summaries with citations and follow-up questions.
- Administration Console
- Django admin pages for SearchProviders, AIProviders, query transforms, users, groups, and API tokens, plus a Log Viewer that tails Django and Celery logs without shell access.
- REST API and Swagger
- Run synchronous or asynchronous federated searches over a documented REST API with an OpenAPI schema, so other applications and agents can consume ranked results and RAG answers.
- Microsoft 365 and Ollama
- Connect Microsoft 365 through OAuth2 so people search as themselves, and enable the preloaded Ollama embeddings provider to run relevancy re-ranking on your own hardware.