Onyx enterprise search

Onyx is an open-source platform for AI search across your own data — 31,900+ stars and 4,400 forks on GitHub, and organisations running it in production worldwide. We do not sell it and we are not reselling a licence: we customise it, host it, scale it and tune it around the material you already have. Where something else fits you better, we will use that instead — the work is the integration, not the tool.

Why open source

The mainstream products in this space are Glean, Microsoft Copilot and Coveo. They're closed SaaS. Your data leaves your boundary, your roadmap depends on theirs, and your pricing is whatever they decide it is next year. For most companies that's fine. For regulators, healthcare providers and public-sector organisations with data-sovereignty obligations, it isn't.

Onyx is open core rather than plainly open source: everything outside its enterprise directories is MIT Expat licensed, and the ee/ directories carry a separate commercial licence. Worth knowing before you build on it, and worth checking against what you intend to deploy. You run it where you want, on whichever LLM you choose (OpenAI, Anthropic, or self-hosted via Ollama or vLLM, including fully offline). You can audit which document chunks went into which answer. It brings 40+ connectors, role-based access control, SSO via OIDC, SAML and OAuth2, document-level permissions, an admin UI, and hybrid search with reranking. Initial deployment usually takes a day; the fit to your material is the part that takes longer.

What we add on top

PretaGov has run two Onyx pilots for AU and UK clients. The extensions below are things we needed and built.

Tables and structured data

Standard RAG breaks tables. Rows and columns get split into fragments, so questions like "what was total revenue last year" return partial answers or no answer at all. We added a parallel SQL pathway. Onyx still does the semantic search; alongside, the question hits a generated SQL query that computes the answer across the full table. End users get a real number instead of a list of links.

PDFs that aren't print-ready

Annual reports, investigation summaries and other regulatory documents have multi-page tables, footnotes and inconsistent layouts. Standard PDF-to-text tools miss those. We swapped Onyx's PDF extractor for one based on PyMuPDF layout analysis, with cross-page table merging. For Excel inputs we added named-table parsing and column unpivoting.

Data sources that aren't documents

Beyond Onyx's 40+ shipped connectors, we've added PowerBI dashboard scraping, table extraction from csv / xlsx / pptx files embedded in web pages, and sitemap improvements that discover linked files (not just linked pages). Our DiscoveryConnector interface makes crawl pruning correct rather than approximate.

Scale-to-zero hosting

For workloads that aren't 24/7, we run Onyx in a scale-to-zero configuration on Fly.io. The single Onyx worker is split into eight specialised workers, each suspended when idle. Cost between requests is zero.

In production

Regulator complaints and oversight. PowerBI dashboards, annual reports, media releases, investigation summaries. A user asks how many complaints were upheld in 2024 and gets a number with citations, not a list of links.

Facing the public instead — answers for citizens on your own website, from your published content? That is Public AI answers.

One of four things we do under AI integration — alongside public answers, ranked search and process automation.