# swirlai/swirl-search

AI Search & RAG Without Moving Your Data. Get instant answers from your company's knowledge across 100+ apps while keeping data secure. Deploy in minutes, not months.

Repository: https://github.com/swirlai/swirl-search
Canonical: https://ross.abutalabs.com/products/swirl-search
Homepage: https://swirlaiconnect.com/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: search, search-engine, federated-query, federated-search, ai-search, bigquery, large-language-models, relevancy, metasearch, django, gpt, python, rag, unified-search, retrieval-augmented-generation
Last push: 2026-08-25T19:06:21+00:00

## Health v2 (maintenance only)
Score: 92/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 77, longevity 100
- inputs: {"age_days": 1609, "days_push": 8, "days_rel": 72, "gap_med": 33.0, "n_releases_24m": 15}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3043, forks 286 (observed 2026-08-28T04:07:39.408204+00:00)

## What it is
SWIRL is an open-source federated AI search and RAG platform that queries enterprise apps and data sources live, re-ranks results with a hybrid keyword/embedding/cross-encoder pipeline, and returns LLM-generated answers with citations - without copying data into a vector database. It is self-hosted via Docker/Kubernetes, enforces source-level permissions, and offers an MCP server so AI agents can search through it.

## Use cases
- search across all company apps without moving data
- get AI answers with citations from internal knowledge
- federated enterprise search with source permissions enforced
- add RAG to company data without building a vector database or ETL
- let AI agents search enterprise data via MCP
- self-hosted AI search over Microsoft 365, SharePoint, and cloud file stores

## When to choose
- you need secure, permission-aware search across many SaaS and enterprise sources
- you want to avoid standing up and governing a vector database or ETL pipelines
- you need self-hosted or air-gapped deployment with your own LLM choice
- you want agents (via MCP) to query company knowledge safely

## When to avoid
- you need to index and deeply analyze a static corpus offline rather than query live sources
- you want a lightweight embeddable search library rather than a deployable service
- you need features like the three-pass reranker, canonical answers, or the full 150+ connector library that are Enterprise-only

## Facets
- artifact type: application
- maturity: active
- function: search-engine, rag, llm-inference, api-framework, mcp, self-hosted
- domain: large-language-models, erp, developer-tools, self-hosted
- platform: python, self-hosted, cloud
- tags: federated-search, metasearch, enterprise-search, no-vector-db, live-query, permissions-aware, django, connectors, ai-search, rag-without-etl, search, retrieval-augmented-generation, docker, kubernetes, web-server

## Member repositories
- swirlai/swirl-search (main) score 92

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:39.408204+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:29:12.647406+00:00, confidence not recorded.
  - readme: https://github.com/swirlai/swirl-search (fetched 2026-08-28T04:07:39.408204+00:00, sha 020507278486)
  - homepage: https://swirlaiconnect.com/ (fetched 2026-08-29T09:44:24.103236+00:00, sha e81648d8f811)
  - site_page: https://docs.swirlaiconnect.com/ (fetched 2026-08-29T09:44:24.114636+00:00, sha 69786fdd80f5)
  - site_page: https://docs.swirlaiconnect.com/claude-plugin (fetched 2026-08-29T09:44:24.146201+00:00, sha 9b5b96b9c4cb)
  - site_page: https://swirlaiconnect.com/releases (fetched 2026-08-29T09:44:24.112540+00:00, sha 3c7f78428f41)
  - site_page: https://swirlaiconnect.com/faq (fetched 2026-08-29T09:44:24.116761+00:00, sha 2edbcc969802)
  - site_page: https://swirlaiconnect.com/pricing (fetched 2026-08-29T09:44:24.118865+00:00, sha f998ddb3dd28)
- Data as of 2026-08-30T08:39:29.467469+00:00.