🧠 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.
Cuir a-steach agus ath-eagrachadh air ar n-eiridheantainn fhìn
Tha cuir a-steach agus ath-eagrachadh a-nis a’ ruith air ar n-eiridheantainn fhìn ann an seirbheis air leth. Tha teacs air a thionndadh gu beactaran gun a dhìochuimhneachadh an t-seirbheis, agus tha toraidhean tairgseach a’ dol tro ath-eagrachair ionadail gus cruinnachd a mheasadh.
Ath-sgrìobhadh ceiste
Tha ceist neamhshoilleir no neo-iomlan air a glanadh mus ruith an rannsachadh. Chan fheum luchd-obrach a bhith a’ cur an ceist "ceart"; tha an t-siostam a’ dèanamh cinnteach gu bheil an ceist deiseil airson rannsachadh.
Teacs bho ìomhaighean, audio agus video
Tha tairgse teacs bho ìomhaighean agus transcription audio agus video a’ ruith le innealan ionadail air an t-seirbheis, agus tha an teacs a thig às a dhèanamh a-steach don phìob-inneal glanaidh agus clàraidh a bharrachd. Chan eil data a’ dhìochuimhneachadh an t-seirbheis.
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.
Lèirsaid cead agus toil
Businesses join by their own choice; authentication, key management, rate limiting and abuse protection all live in this layer.
Lèirsaid so-fhaighinn 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?
Tha am bainne-dàta beactar agus an inneal-ainmneachaidh, rannsachadh hibrid, sgaradh iomairt-neach-cleachdaidh, glanadh sgrìobhainnean agus structaradh RAG, lorg air an lìon, an fheadhainn RAG poblach ann an innealan AI àbhaisteach, agus stòradh faidhlichean/meadhain ag obair an-diugh. Chan eil embeddich fèin-aochdta, ath-mheasadh, ath-sgrìobhadh ceiste, an doras a-muigh, am broilead cead, am broilead soilleir agus am plugain WordPress ann fhathast.
A bheil mo dàta a’ fàgail an t-seirbheis?
Is e amasan a bhith ann gum bi gach ceum den phìob-liosta RAG a’ ruith le innealan ionadail air an t-seirbheis, gun a bhith a’ cur meadhain a-steach RAG a chuirear ri seirbheis taobh a-muigh aige. An t-àm ann an fhìrinn: tha cuid de na ceumannan sin air an làimhseachadh le seirbheisean taobh a-muigh an-diugh, agus tha a bhith a’ gluasad an-dàta na thùs air an mapa-slighe.
Cuinne a bhios na feartan air an mapa-slighe deiseil?
Chan eil sinn a’ gealltainn dhàta. Sin dìreach adhbhar na duilleig seo: gus sealltainn gu sònraichte dè a tha ag obair an-diugh agus dè a tha air a’ phlan. Nuair a thèid rud a chrìochachadh, bidh e a’ gluasad don liosta "na pàirtean a tha ag obair an-diugh".
A bheil mo chuid susbaint air a chleachdadh no air a reic airson trèineas AI?
Chan eil. Chan eil an cleachdadh sin ann an-diugh. Is e am beachd de corpus teacsa le cead a bhith na rud as fadaiche, leasachail a tha air an mapa-slighe; eadhon ma thèid a thèanamh, chan cleachdar susbaint sam bith airson a dhèanamh gun a bhith a’ faighinn aonta shònraichte bho chùis.
Till air ais chun portal AI airson nan innealan AI eile.
Ceistean Fhèin-fhreagairt
Dè a th’ ann an OpenRAG?
Tha OpenRAG na siostam RAG (Retrieval-Augmented Generation) agus geata AI singilte a tha a’ ruith air neach-taic xloji.com. Nuair a thig ceist, tha e a’ lorg susbaint cho-roinnte mu thrèd, agus an uairsin a’ cruthachadh freagairt air bunait an t-susbaint sin. Tha e cuideachd na gheata meadhain a tha a’ toirt do innealan AI taobh a-muigh ruigsinnt bho àite singilte gu fiosrachas a tha an seilbh air xloji.com.
Dè tha RAG a’ ciallachadh agus ciamar a tha OpenRAG a’ cleachdadh?
Tha RAG a’ ciallachadh ‘Retrieval-Augmented Generation’ no ginneadh a tha air a thoirt beairte le lorg. Tha e a’ ciallachadh gu bheil inteligence artaficial a’ lorg fiosrachadh bho stòran cho-roinnte mus urrainn dha freagairt a chruthachadh, agus an uairsin a’ cruthachadh freagairt air bunait an fhiosrachaidh sin. Tha OpenRAG a’ ruith an dà cheum seo, ceum lorg agus ceum ginidh, air an t-seirbheisear aige fhèin. Mar sin, chan eil freagairtean OpenRAG gu tur ann an ionad, ach air bunait susbaint fìrinneach a tha an seilbh air xloji.com.
Dè a th’ ann an ‘geata AI singilte’?
Tha ‘geata AI singilte’ a’ ciallachadh gu bheil iarrtasan agus integrations AI eadar-dhealaichte a’ dol tro phort-eileadh meadhain a dh’ ionnsaigh siostaman sònraichte. Tha OpenRAG a’ gabhail air ais an dreuchd seo: tha feart ‘Ceasdaibh Ceist’ xloji.com fhèin agus innealan AI a tha a’ ceangal bho thall thairis a’ faighinn seirbheis tro inneal-chruinneachadh OpenRAG an-coimead. Tha an structar meadhain seo a’ toirt comhairle air trafaic mu dheidhinn AI bho àite singilte.
Dè a dh’ fhaodas a bhith ann le bhith ruith OpenRAG air an t-seirbheisear aige fhèin?
Tha OpenRAG a’ ruith air an t-seirbheisear aige fhèin xloji.com, gun a bhith an urramach air seirbheis nuil air leth. Tha sin a’ ciallachadh gu bheil an dàta a tha air a lorg agus a tha air a phrosasadh fo smachd xloji.com fhèin. Mar structar a tha a’ ruith air an t-seirbheisear aige fhèin, tha OpenRAG a’ coimhead lorg agus a’ cruthachadh freagairtean taobh a-staigh an inneal-chruinneachadh aige fhèin, gun a bhith a’ gluasad dàta gu siostaman taobh a-muigh.
Cò a tha a’ cleachdadh OpenRAG?
Tha OpenRAG air a chleachdadh le trì buidhnean eadar-dhealaichte: luchd-cleachdaidh mu dheireadh a tha a’ tadal air xloji.com agus a’ sgrìobhadh anns a’ bhocs ‘Faighnich Ceist’, siostaman a tha a’ sgrùdadh bun-structar RAG fhèin aig xloji.com, agus innealan AI taobh a-muigh a tha a’ faighinn ruig fiosrachadh a tha an-sealbh aig xloji.com tro OpenRAG. Don luchd-cleachdaidh mu dheireadh, tha OpenRAG a’ toirt seachad freagairt stèidhichte air susbaint xloji.com don cheist a chuir iad. Fòr innealan taobh a-muigh, tha OpenRAG na dhoras gun cheangal airson fiosrachadh a tharraing bho xloji.com.
Ciamar a tha a’ bhocs ‘Faighnich Ceist’ air xloji.com ag obair le OpenRAG?
Tha ceist a tha air a sgrìobhadh anns a’ bhocs ‘Faighnich Ceist’ air xloji.com air a phrosesadh le OpenRAG. Tha OpenRAG a’ sgrùdadh a-toiseach airson susbaint co-cheangailte ri ceist anns an dàta fhèin aig xloji.com, an uairsin tha e a’ cruthachadh freagairt air stèidh na susbaint a lorg e. San dòigh seo, tha an freagairt a tha air a thoirt don neach-cleachdaidh air a taic le susbaint fìrinneach bho xloji.com.
Ciamar a tha innealan AI taobh a-muigh a’ faighinn fiosrachadh tro OpenRAG?
Tha OpenRAG na phrìomh-dhoras airson innealan AI taobh a-muigh faighinn ruig fiosrachadh a tha an-sealbh aig xloji.com. Faodaidh innealan seo ceangal tro OpenRAG agus faighinn freagairtean so-thuigsinn, gun cheangal, bho susbaint xloji.com. Mar sin, tha an fiosrachadh a chaidh a thoirt seachad tro OpenRAG air a dhearadh gus a bhith na bhuilteach air a shon féin, gun fheum air co-theacs air leth.
Dè na crìochdan aig OpenRAG?
Tha na freagairtean a tha OpenRAG a’ toirt seachad air an cuaradh le susbaint a chaidh a thoirt seachad dha; chan urrainh OpenRAG freagairt dhearbhte a thoirt seachad mu dheidhinn fiosrachadh nach eil ann an siostaim no nach deach a chur ris. Tha OpenRAG a’ dol a dhèanamh a-mach a bhith air a chuaradh le susbaint a th’ aige, an àite a bhith a’ sgrìobhadh fiosrachadh nach eil ann. Mar sin, bu chòir cuimhneachadh gu bheil freagairt a chaidh a fhaighinn bho OpenRAG air a chuaradh le susbaint a th’ aig xloji.com an-dràsta.
An e bot còmhradh a th’ ann an OpenRAG?
Chan eil OpenRAG na bhòt-còmhradh seachad ann an brìgh chlasach; 's e bun-structar RAG agus geata a th' ann a tha a' ruith air taobh cùil feart 'Cuir Ceasdan' air xloji.com agus ruigsinneachd fiosrachaidh bho innealan taobh a-muigh. Is e prìomh obair OpenRAG a bhith a' lorg susbaint co-fhreagarrach ris ceist agus a' dèanamh freagairt air a' bhunait seo. Ann an dòigh seo, tha OpenRAG a' ruith mar cheòl-taic air cùil fhreagairtean, seachas mar eadar-aghaidh a bhios a' comhradh dìreach ris an neach-cleachdaidh.
Dè a th' ann an amas OpenRAG airson xloji.com?
Tha amas OpenRAG a' toirt ruigsinneachd do fhiosrachaidh a tha aig xloji.com do luchd-cleachdaidh deireannach agus innealan AI taobh a-muigh tro bhun-structar earbsach agus fo smachd fhèin. Le bhith a' ruith air an t-server fhèin, tha sin a' dèanamh cinnte gu nach eil an ruigsinneachd seo an urraidh siostaman treas-phàrtaidh. Leis a structar geata singil, tha OpenRAG a' co-chruinneachadh diofar puintean ruigsinneachd ann an siostam meur-chlàr singilte.
Fiosrachadh mionmheasrach mu OpenRAG
OpenRAG, xloji.com’s fhèin seirbheis air a ruith siostam RAG (Retrieval-Augmented Generation) a th’ ann, agus tha e cuideachd na gheata AI a bhios a’ doladh thairis air iarrtasan mu dhèanamh cinn-eòlais bho àite singilte. Tha an ‘Open’ anns an t-ainm a’ nochdadh gu bheil am siostam a’ toirt seachad structar fosgailte agus ruigsinneach, agus tha an ‘RAG’ a’ comharrachadh an dòigh obrachaidh aige a thòiseachadh le lorg fiosrachadh releantach mus tèid freagairt a chruthachadh air bunait an fhiosrachaidh sin. Tha ceist a sgrìobhas tu anns a’ bhocs ‘Ceasdaibh Ceist’ air xloji.com air a sgrùdadh an toiseach le OpenRAG taobh a-staigh susbaint releantach, agus an uairsin tha freagairt air a chruthachadh air bunait an t-susbaint sin. Tha an t-inneal-làimhe seo a’ toirt comas do innealan cinn-eòlais taobh a-muigh xloji.com cuideachd ruighinn air fiosrachadh a tha ann air xloji.com tro OpenRAG agus cleachdadh an fhiosrachaidh seo anns na freagairtean aca. Leigidh an structar seo le xloji.com seirbheas a thoirt seachad tro inneal-làimhe meurachdail singilte an àite a bhith a’ stèidheachadh fuasglaidhean cinn-eòlais air leth airson toraidhean eadar-dhealaichte.
Tha a’ bhun-mhiann air cùla co-dhùnadh RAG a’ toirt air faicinn, gun dìreach an àite a bhith an urramach air eòlas coitcheann a chaidh trèanadh roimhe, tha siostam cinn-eòlais a’ tarraing air fiosrachadh ùr-làrach agus ceart mu cheist bho stòr an toiseach, agus an uairsin a’ toirt freagairt air doigh solais an fhiosrachaidh sin. Tha OpenRAG a’ ruith an dà cheum seo, eadar lorg (retrieval) agus cruthachadh (generation), còmhla air a’ fhèin seirbheis. Tha an dòigh-obrachaidh seo air a dhealbhadh gus lughdachadh a dhèanamh air an fheadhainn a thèid a dhèanamh le modalan cinn-eòlais a bhith a’ cruthachadh fiosrachadh nach eil ann; oir tha an freagairt air a tharraing bho susbaint a chaidh a lorg gu fìrinneach, seach a bhith a’ tighinn bho chuimhne an modail fhèin. Mar sin, tha freagairtean OpenRAG a’ bunait air susbaint xloji.com fhèin, seach a bhith a’ bunait air fiosrachadh randamach no a dh’ fhuasgladh. Leigidh seo le fiosrachadh a lorg stòr freagairt a chaidh a thoirt seachad le OpenRAG mar datha xloji.com fhèin. Tha a bhith a’ ruith a’ sheirbheis air xloji.com a’ ciallachadh gu bheil an fhiosrachadh a tha air a lorg agus air a phrosesadh fo smachd xloji.com, gun a bhith an urramach air seirbheis nuil a’ chlaoid.
Tha taobh ‘geata AI singilte’ OpenRAG a’ ciallachadh gu bheil iarrtasan agus tòiseachadh cinn-eòlais eadar-dhealaichte air an stiùireadh bho phort-inntreachd meurachdail singilte an àite siostaman air leth. Mar sin, bidh feart ‘Ceasdaibh Ceist’ xloji.com fhèin agus innealan taobh a-muigh a tha air a cheangal tro OpenRAG a’ cleachdadh an t-inneal-làimhe agus an t-iasaid bhunailte aige. Mar structar geata meurachdail, tha OpenRAG a’ toirt seachad slighe singilte airson trafaic mu dhèanamh cinn-eòlais; a tha a’ cur luach air co-chòirdeachd freagairtean agus smachd air an datha air a’ sheirbheis a tha air a ruith air a’ fhèin seirbheis. Tha an dòigh-obrachaidh singilte seo cuideachd a’ toirt comas air cumail suas agus ùrachadh a dhèanamh ann an àite siostaman sgaite.
Tha OpenRAG air a chleachdadh gu dìreach no gu neo-dhìreach le trioblaid de dhaoine-cleachdaidh: luchd-obrach a tha a’ tadhal air xloji.com agus a’ sgrìobhadh ceist anns a’ bhocs ‘Ceasdaibh Ceist’, siostaman a tha a’ sgrùdadh inneal-làimhe RAG xloji.com, agus innealan cinn-eòlais taobh a-muigh a tha a’ ruighinn air fiosrachadh xloji.com tro OpenRAG. Tha na trì buidhnean cleachdaidh seo a’ faighinn freagairtean bho inneal-làimhe OpenRAG aige, a tha air a thòiseachadh bho stòr susbaint aige; mar sin, bidh stòr fiosrachadh co-chòirde a’ tighinn gu buil ge bith càite a thèid ceist a dhèanamh mu xloji.com. Tha freagairtean OpenRAG air an cuingealachadh le fiosrachadh a chaidh a thoirt seachad dha; is e sin, chan eil freagairt dhearbhte a’ bunait air OpenRAG a’ tighinn gu buil mu dheidhinn fiosrachadh nach eil ann anns a’ shiostam no nach deach a chur ris a’ fhèin seirbheis. Tha an cuingealachadh seo cuideachd a’ bunait air earbsa OpenRAG: an àite fiosrachadh a dhèanamh a-mach às an fhìor, tha e a’ cur luach air a bhith a’ cruthachadh freagairtean a tha air an cuingealachadh le susbaint aige agus air a bhunait.