🧠 OpenRAG
OpenRAG is a RAG platform that cleans and structures a business's content, indexes it on our own server, and makes it queryable in natural language by external AI systems through a single permissioned door.
Vectors are not published; ACCESS is opened.
This is a product introduction, not a list of promises. Below, the parts that work today and the parts still on the roadmap are marked separately. No dates are promised for roadmap items; as they are completed they move into the "working today" section.
The core principle
OpenRAG's entire design rests on a single sentence: vectors are not published, access is opened. What is given to the outside is not a copy of the content, but the right to ask questions of it.
A vector is an engine, not a product
A vector is specific to the model that produced it; another AI cannot consume it as-is. It also inflates the text many times over. That is why vectors are never handed out.
What is opened is query access
A question comes in; an answer and the source passages it rests on go back out. The full content is never copied, distributed or downloaded.
Permission belongs to the business
The business decides which content becomes externally queryable; participation is opt-in. The permission and consent layer is on the roadmap.
How it works — target architecture
The cleaning, structuring, indexing and search links of this flow work today. Self-hosted embedding with reranking, and the single door to the outside world, are still on the roadmap; each step below states which is which.
Sources are collected
xloji sites, uploaded documents and crawled web pages are the content sources. (A WordPress plugin and a public ingestion API are on the roadmap.)
It is cleaned and structured
Messy content is cleaned, split into meaningful chunks and brought into a form RAG can use. This step works today.
It is indexed and searched
Chunks are turned into vectors and written to pgvector; they are found via an HNSW index and hybrid search (keyword + vector). This step works today; adding the reranking link is on the roadmap.
It is reached through a single door
External AI systems ask in natural language through a single permissioned door; answers and source passages come back, not the vectors themselves. This door does not exist yet; it is on the roadmap.
Parts that work todayLive
The following are running on the platform today and are already used by the AI tools on xloji.com.
pgvector + HNSW index
Content chunks are stored as vectors in the database and searched with an HNSW index over cosine similarity.
Hybrid search
Keyword search and vector search run together; the two result lists are fused into a single ranking. This finds both exact term matches and semantically close passages.
Multi-tenant isolation
Every business, site and author has its own RAG. A query runs only against its own tenant's index; data never mixes across tenants.
Document cleaning and RAG structuring
Uploaded documents are cleaned, chunked and structured for RAG. The Data Structuring tool on the platform is the user-facing face of this pipeline.
Web crawling
A site's pages are crawled, their content is extracted and passed through the same cleaning and indexing pipeline.
Public RAG feed in custom AI tools
A custom AI tool you create can be fed from a public RAG source; the tool grounds its answers in that knowledge base.
File and media storage
Files and media live in object storage — not in the vector store. The RAG record points back to the original file.
Ukufaka nokuhlela kabusha kumaseva ethu
Ukufaka nokuhlela kabusha manje kwenziwa kumaseva ethu esevisi ehlukile. Umbhalo ugcinwa njengezivekhetha ngaphandle kokuphuma kweseva, futhi imiphumela yomgomo iyadlula kusihleli sokuhlela kabusha esisebenzisa ngaphakathi ukuze kuthuthukiswe ukunemba.
Ukuphinda kubhalwa kwembuzo
Umbuzo ongacacile noma ongagcwele ugcinwa uhlelekile ngaphambi kokuthi usesho luqalwe. Abasebenzisi akudingeki ukuthi babhale "umbuzo olungile"; uhlelo lwenze imbuzo ilungele ukuseshwa.
Umbhalo kusuka ezithombeni, kumsindo nakumavidiyo
Ukukhipha umbhalo ezithombeni nokubhala komsindo namavidiyo kwenziwa ngamathuluzi asendaweni kuseseva, futhi umbhalo ovelayo ungena esigubheni esifanayo sokuhlanzwa nokufakwa kwamanani. Idatha ayiphumi esevani.
What is on the roadmapPlanned
The following do not exist yet. They are planned and designed items; this page does not present them as if they were already built.
The single door: MCP / API gateway
A single gateway that external AI systems can query in natural language. An AI that connects once reaches every participating business through the same door.
Isiqubulo nokuvumela kwesiqu
Businesses join by their own choice; authentication, key management, rate limiting and abuse protection all live in this layer.
Isiqubulo sokubonakala kwe-AI
Making content machine-discoverable via llms.txt, /.well-known/ and schema.org markup, plus registration in MCP registries. There is no magical auto-discovery; visibility is built this way.
WordPress plugin
A plugin that lets a WordPress site send its content to OpenRAG and get back an endpoint that AI systems can query.
We do not give dates for these items. When an item is finished it moves up into the "parts that work today" list, and this page is updated accordingly.
🔒 Data does not leave the server
OpenRAG's goal is to run every step of the RAG pipeline on our own server with local, free tools: embedding, reranking, text extraction from images and audio transcription. No media entering RAG is sent to an external service — this is a non-negotiable design rule.
The honest status: some of these steps are still performed by external services today. Moving them in-house is the first item on the roadmap.
How growth is planned
Vector search gets more expensive as content grows. OpenRAG's design limits that cost from the start in three ways.
There is no single giant index
Each business has its own small index and a query runs only against that index. Even as total content grows, the area a single query scans stays small; search cost scales with that one business's size, not with the platform's total.
Vector size and precision are adjustable
Using shorter vector dimensions and more compact number formats, the same content takes up markedly less space. These choices are built into the design from the start so that no expensive migration is needed later.
Cold tier
The index of a business that has not been queried for a long time is moved to cheap persistent storage and pulled back when a query arrives for it. That way only currently active content stays in fast memory. This tier is on the roadmap.
Frequently Asked Questions
What exactly is OpenRAG?
It is a RAG platform that cleans and structures a business's content, indexes it on our own server, and makes it queryable in natural language by external AI systems through a single permissioned door. It is not a chat tool; it is the knowledge layer that runs underneath such tools.
Do you hand my vectors or my content to anyone else?
No. The core principle is: vectors are not published, access is opened. What goes outside is not a copy of the content but the right to ask it questions; only the supporting passages come back with the answer. On top of that, the business decides which content becomes externally queryable.
What can actually be used today?
I-database yomvikeli ne-index, ukusesha okuhlanganisiwe, ukuhlukaniswa kwamakhasimende amaningi, ukuhlanzwa kwemibhalo nokwakhiwa kwe-RAG, ukucindezela kuwebhu, ukudla kwesizindalwazi se-RAG esizinkambweni ze-AI ezizimele, nokugcinwa kwamafayela/kwemidiya konke kusebenza namuhla. Ukufakwa kwe-embedding okuzizimele, ukuhlela kabusha, ukubhala kabusha imibuzo, umnyango wangaphandle, isihloko sokuvumela, isihloko sokubonakala, neplagi ye-WordPress akekho okwamanje.
Ingabe idatha yami iyashiya iseva?
Inhloso ukuthi wonke umzuzu we-pipeline ye-RAG usebenze ngamathuluzi asendaweni kuseva, ngaphandle kokuthi noma yikuphi okungena ku-RAG kuthunyelwe kwisevisi yangaphandle. Isimo esiqinisekile: ezinye zalezi zinyathelo ziphathwa izinsizakalo zangaphandle namuhla, nokuzihambisa ngaphakathi kuyinto yokuqala ohlwini lwemigwaqo.
Nini lapho izici zohlelo lokusebenza zizolungile?
Asithembisi izinsuku. Yilokho okwenza le nkhomba ibe yinto ebalulekile: ukukhomba ngokuhlukile okusebenza namuhla nokuhleliwe. Lapho into iqediwe, iyathuthela ohlwini lwezinto “ezisebenza namuhla”.
Ingabe okuqukethwe kwami kusetshenziswa noma kuthengiswa ukuqeqesha i-AI?
Cha. Akukho ukusetshenziswa okunjalo namuhla. Umqondo we-corpus yombhalo ovunyelwe yinto ekude kakhulu, enesisindo esikhulu ngomthetho ohlwini lwemigwaqo; ngisho noma kwenzeka, akukho okuqukethwe okuzosetshenziswa kuona ngaphandle kwemvume ecacile yebhizinisi.
Buyela emuva ku-portal ye-AI ukuze uthole amanye amathuluzi e-AI.
Imibuzo Evame UkuBubuzwa
I-OpenRAG yini?
I-OpenRAG iyisistimu ye-RAG (Retrieval-Augmented Generation) esisekelwe kweseva ye-xloji.com, futhi ingumnyango owodwa we-AI. Lapho umbuzo ufika, iyathola kuqala okuqukethwe okufanele, bese ikhiqiza impendulo esekelwe kokuqukethwe. Ngasikhathi sinye, isebenza njengomnyango omaphakathi ovumela amathuluzi a-AI angaphandle ukuthi afinyelele ulwazi olungelwe yi-xloji.com kusuka endaweni eyodwa.
Kusho ukuthini i-RAG, futhi i-OpenRAG isisebenzisa kanjani?
I-RAG isho 'Retrieval-Augmented Generation' - okusho ukukhiqizwa okugcwele ukutholwa; i-AI isesha ulwazi emithonjeni efanele ngaphambi kokukhiqiza impendulo, bese ikhiqiza impendulo esekelwe kulolo lwazi. I-OpenRAG isebenza ngokuhlanganisa lezi zinyathelo ezimbili, ukusesha nezinyathelo zokukhiqiza, kweseva yayo. Ngakho, izimpendulo ezihlala zinikezwa yi-OpenRAG azikho ngokungenaphi, kodwa zisekelwe kokuqukethwe kwangempela okungelwe yi-xloji.com.
Kusho ukuthini 'umnyango owodwa we-AI'?
Umnyango owodwa we-AI usho ukuthi izicelo ze-AI ezahlukahlukene nokuhlanganiswa kwazo kuphelelwa ngomnyango owodwa omaphakathi, esikhundleni samasistimu ahlukahlukene. I-OpenRAG ithatha leli bhrole: kokubili isici se-'Buzwa Umbuzo' se-xloji.com, kanye namathuluzi e-AI axhunyekwe ngaphandle, usebenzela ngesakhiwo se-OpenRAG esifanayo. Lolu sakhiwo esimaphakathi sivumela ukuphathwa kwesikhathi esisodwa esihlobene ne-AI.
Yini umehluko wokuthi i-OpenRAG isebenze kweseva yayo?
I-OpenRAG isebenza kweseva ye-xloji.com, inganciki kunoma yisiphi isevisi sefusha sesithathu. Lokhu kusho ukuthi idatha eseshiwe nencishiswe ihlala ilawulwa yi-xloji.com. Njengesisakhiwo esisebenza kweseva yayo, i-OpenRAG iqhuba ukusesha nokukhiqizwa kwempendulo ngaphakathi kwesakhiwo sayo, ngaphandle kokudlulisela idatha kumasistimu angaphandle.
Ngubani osezisebenzisa i-OpenRAG?
I-OpenRAG isetshelwa iziqembu ezithathu ezahlukene: abasebenzisi bokugcina abathayela umbuzo ebhokisini elithi 'Buza Umbuzo' ku-xloji.com, izinhlelo ezibuza imvelo ye-RAG ye-xloji.com, kanye nezinsiza zokwenziwa kwamasiko zangaphandle ezithola ulwazi olungelwe-xloji.com nge-OpenRAG. Kumsebenzisi wokugcina, i-OpenRAG inikeza impendulo esekelwe kuzinto ze-xloji.com. Ngezinsiza zangaphandle, i-OpenRAG iyisango elingena konke elisetshelwa ukuthola ulwazi ku-xloji.com.
I-OpenRAG isebenza kanjani nebhokisi elithi 'Buza Umbuzo' ku-xloji.com?
Umbuzo ofakwe ebhokisini elithi 'Buza Umbuzo' ku-xloji.com uyacutshungulwa yi-OpenRAG. I-OpenRAG iyacwaninga kuqala ngokuqukethe okufanele umbuzweni ngaphakathi kwedatha ye-xloji.com, bese yakha impendulo ngokusekelwe kulokho okuwutholile. Ngalesi sizathu, impendulo enikezwe umsebenzisi isekelwe ngokuqukethe kwangempela kwe-xloji.com.
Izinziza zokwenziwa kwamasiko zangaphandle zithola ulwazi kanjani nge-OpenRAG?
I-OpenRAG isebenza njengesango elilodwa lokudlula, elivumela izinsiza zokwenziwa kwamasiko zangaphandle ukuthi zifinyelele ulwazi lwe-xloji.com. Lezi zinsiza zingaxhumana nge-OpenRAG futhi zithole izimpendulo ezizimele, ezingaqondakala ngokwazo kusuka kuzinto ze-xloji.com. Ngalesi sizathu, ulwazi olunikezwe nge-OpenRAG luyilwe ukuba nesincazelo ngokwalo ngaphandle kwesidingo sesimo esihlukile.
Yiziphi imingcele ye-OpenRAG?
Izimpendulo ezinikezwe yi-OpenRAG zivikelekile ngokuqukethe okunikezwe kuyo; i-OpenRAG ayikwazi ukukhiqiza impendulo eqinisekile mayelana nolwazi olungatholakali noma olungafakwanga esistimeni. I-OpenRAG ihlose ukuhlala ngaphakathi kokokuqukethe okuyitholayo, kunokudala ulwazi olungakhona. Ngalesi sizathu, kufanele kukhunjulwe ukuthi impendulo etholwe ku-OpenRAG ivikelekile ngokuqukethe okwamanje okwaba nakho i-xloji.com.
Ingabe i-OpenRAG iyisikhathi sokuxoxa?
I-OpenRAG ayisiyo yona umngani wokuxoxa ozimele ngokusho kwesiko; iyisisekelo se-RAG kanye nesango esisebenza ngemuva kwesici esithi 'Buza Umbuzo' se-xloji.com nokufinyelela kwemithombo yezindaba yangaphandle. Umsebenzi oyinhloko we-OpenRAG ukuthola okuqukethwe okufanele ukuphendula umbuzo, bese ukhiqiza impendulo ngalokho kokuqukethwe. Ngalesi simo, i-OpenRAG isebenza njengesendlalelo ngemuva kwezimpendulo, kunokuba isebenzisane ngqo nomsebenzisi.
Yini inhloso ye-OpenRAG ye-xloji.com?
Inhloso ye-OpenRAG ukunikeza i-xloji.com ukufinyelela olwazi oluthembekile nolawulwa yiyo, kubasebenzisi abajwayele futhi kumathuluzi obuhlakani bangaphandle, ngesisekelo esithembekile. Ukusebenza kweseva yayo kwenza lolu kufinyelela kwenzeke ngaphandle kokuncika ezinhlelweni zangaphandle. Ngenxa yesakhiwo sayo sesango elilodwa, i-OpenRAG ihlanganisa izindawo ezihlukile zokufinyelela endaweni eyodwa ephakathi.
Imininingwane ephelele mayelana OpenRAG
I-OpenRAG, iyisistimu ye-RAG (Retrieval-Augmented Generation) esebenza kweseva ye-xloji.com, futhi futhi ingongoma ye-AI eyidlula izicelo ezihlobene ne-AI kusuka kwenye indawo. Igama elithi 'Open' liguqula ukuthi lesi sistimu sinikeza isakhiwo esivulekile futhi esifinyelelekayo, kanti igama elithi 'RAG' likhomba indlela yokusebenza yokuthola okuqukethe okufanele okokuqala lapho kukhiqizwa izimpendulo, bese kwakhiwa izimpendulo ngokusekelwe kuleyo data. Umbuzo othatyathwa ebhokisini elithi 'Buza Umbuzo' ku-xloji.com ucwaningwa kuqala ngumhlaba we-OpenRAG ngaphakathi kwokuqukethwe okufanele, bese kusetshenziswa lokuqukethwe olutholwe ukukhiqiza impendulo. Isakhiwo esifanayo sivumela amathuluzi e-AI ngaphandle kwe-xloji.com ukuthi afinyelele ulwazi olungaphakathi kwe-xloji.com nge-OpenRAG futhi alusebenzise lolo lwazi ezimpendulweni zabo. Lesi sakhiwo sivumela i-xloji.com ukuthi inikeze izinsizakalo kusuka kusakhiwo oluyingqayizivele esikhundleni sokwakha izixazululo ze-AI ezihlukile zemikhiqizo ehlukile.
Inqubo eyisisekelo ye-RAG ukuthi, esikhundleni sokuthembela olwazini olujwayelekile olufundisiwe ngaphambilini, uhlelo lwe-AI kufanele likhethe kuqala okuqukethwe kwesikhathi esifanayo futhi okunembile kusuka emthonjeni, bese liphendula ngokusekelwe kulolu lwazi. I-OpenRAG isebenza ngokuhlanganisa lezi zinyathelo ezimbili, okungukuthi, ukuthola (retrieval) kanye nokukhiqiza (generation), kweseva yayo. Le ndlela ihloswe ukunciphisa umoya we-AI wokukhiqiza ulwazi olungakhona; ngoba impendulo ivela ekukheni okutholwe ngempela, hhayi kumemori yomuzini. Ngalesi sizathu, izimpendulo ezinikezwe yi-OpenRAG azisekelwanga kulwazi olungahambi kahle noma oludlule esikhathini, kodwa kokuqukethwe kwesivumelwano se-xloji.com. Ngakho umthombo wempendulo ekhiqizwa yi-OpenRAG ungagcinwa njengoba eqoshwe njengoba kwenziwe ngedatha ye-xloji.com. Ukuthi iseva ingeyomnikazi we-xloji.com kusho ukuthi idatha eseshelwa futhi eyenziwa igcinwe ngaphakathi kokulawulwa kwe-xloji.com ngaphandle kokuncika kwisevisi yefu yomuntu wesithathu.
Ohlangothi lwe-'single-door AI gateway' lwe-OpenRAG, kusho ukuthi izicelo zokuhlanganiswa ne-AI ezihlukile ziyahambisa kusuka kwenye indawo yokungena esiyingqayizivele esikhundleni samasistimu ahlukene. Ngalesi sizathu, kokubili isici se-'Buza Umbuzo' se-xloji.com kanye namathuluzi angaphandle axhumene nge-OpenRAG, asebenzisa isakhiwo esifanayo kanye nenqubo eyisisekelo efanayo. Njengesakhiwo esiyingqayizivele, i-OpenRAG ivumela ukuthi umgwaqo we-AI udlule endaweni eyodwa, okukhulisa kokubili ukulingana kwezimpendulo nokulawulwa kwedatha njengoba kusebenza kweseva yayo. Le ndlela yokufanela eyodwa ivumela nokuthi ukunakekelwa nokuvuselelwa kwenziwe endaweni eyodwa esikhundleni samasistimu ahlukile. Ngalesi sizathu, i-OpenRAG ayiyona nje ithuluzi eliphendula imibuzo, kodwa iphinde ibe yisendlalelo esihlanganisa wonke amaphoyinti okufinyelela e-AI e-xloji.com.
Kunezintathu iziqembu ezahlukene zabantu abasebenzisa i-OpenRAG ngqo noma ngokungaqondile: abasebenzisi bokugcina abavakashela i-xloji.com futhi bathayipha umbuzo ebhokisini elithi 'Buza Umbuzo', amasistimu aphoqa isakhiwo se-RAG se-xloji.com, namathuluzi e-AI angaphandle afinyelela ulwazi lwe-xloji.com nge-OpenRAG. Lezi ziqembu ezintathu zithola izimpendulo ezivela kusakhiwo se-OpenRAG esifanayo, ezivela echibini lokuqukethwe elifanayo; ngakho, noma kubuzwa mayelana ne-xloji.com kuphi, umthombo olungile wolwazi uyatholakala. Izimpendulo ezinikezwe yi-OpenRAG zixutshwe ngokuqukethwe okunikezwe kuyo; okusho ukuthi akufanele kulindelwe impendulo eqinisekile ku-OpenRAG mayelana nolwazi olungatholakali ngaphakathi kwesistimu noma okungeke kwengezwe kweseva yayo. Lo mkhawulo futhi lususelwa ekuthembekeni kwe-OpenRAG: esikhundleni sokudala ulwazi olungakhona, ihlose ukukhiqiza izimpendulo ezilinganiselwe ngokuqukethwe eliyatholakala.