Semantic search over your documentation - with no LLM at query time.
Zero hallucination. Zero per-query cost. Returns in milliseconds.
Your AI asks a question. Ragionex returns the exact passage - with source, with media, without inventing a single word.
Drop your docs in. We turn the chaos into context your AI can trust.
Send us PDFs, DOCX, Markdown, HTML. Text, images, video - all of it. We turn the messy reality into clean context your AI can use.
Every image, diagram, and recording in your docs is indexed and searchable. Built to find the right answer every time.
POST a question to /v1/knowledge/search. We return the most relevant passages with sources - ready to quote inside your model's context window.
We did the engineering most teams underestimate. Six months of context infrastructure, distilled into one HTTP call. You ship the AI.
Real users phrase things badly. We made it not matter. Same answer, every time.
Every query: instant, predictable, free. No AI bills at runtime. We did the heavy lifting once, upfront. You just ask.
Half your docs are images. Most search engines treat them like decoration. We taught ours to see. Users find answers inside the picture.
Purpose-built retrieval runs entirely on our infrastructure. Your AI gets context faster than it can start generating a single token.
The demo key is indexed on the full VS Code documentation (licensed under CC BY 3.0 US) - every feature, command, setting, screenshot, and video walkthrough is searchable. Shared across testers, rate-limited per IP. Pick a language below, copy the snippet, and run it. See the docs for more info →
import requests response = requests.post( "https://api.ragionex.com/v1/knowledge/search", headers={"X-API-Key": "rgx_knowledge_demo_geXuBZJ5O2GltcG63s1LGijGBwlERGlo"}, json={ "question": "How do I format code in VS Code?", "collection": "vscode-docs", "results": 10, }, ) data = response.json() print(data)
const response = await fetch( "https://api.ragionex.com/v1/knowledge/search", { method: "POST", headers: { "Content-Type": "application/json", "X-API-Key": "rgx_knowledge_demo_geXuBZJ5O2GltcG63s1LGijGBwlERGlo", }, body: JSON.stringify({ question: "How do I format code in VS Code?", collection: "vscode-docs", results: 10, }), }, ); const data = await response.json(); console.log(data);
curl -X POST https://api.ragionex.com/v1/knowledge/search \ -H "Content-Type: application/json" \ -H "X-API-Key: rgx_knowledge_demo_geXuBZJ5O2GltcG63s1LGijGBwlERGlo" \ -d '{ "question": "How do I format code in VS Code?", "collection": "vscode-docs", "results": 10 }'
Ask the way you would in a chat. Full, natural questions return the precise passage; loose keywords miss the meaning and return weak matches.
debug python breakpoints How do I debug Python in VS Code? A focused question gets the sharpest answer. Open-ended prompts like "tell me everything" come back vague - if you need broader coverage, just make several focused calls and combine the results in your app.
Tell me everything about debugging in VS Code How do breakpoints work in VS Code? Let your AI agent break complex, multi-topic questions into simple sub-questions automatically. Each sub-question returns a clean answer; feed them back to your AI as context so it can pull the relevant pieces and craft the final reply.
How do I use Git and configure the terminal in VS Code? How do I use Git in VS Code? How do I configure the terminal in VS Code? Pass "collection": "vscode-docs" - the only collection on this demo key. Tune "results": N to size each response - higher N returns more related passages, so set it to fit your use case. The engine always returns the closest matches it has, so off-topic questions get the nearest VS Code passage, not silence.
How do I edit images in Photoshop? How do I split the editor in VS Code? Be first to try Ragionex with your own documentation.
We're opening Closed Beta soon. Drop your email and we'll notify you when it's your turn.
No spam. Just one email when beta opens.