Get your API key

Knowledge uses a single shared key during Developer Preview. Grab the rgx_knowledge_demo_ key from the Knowledge page - no sign-up required to try it.

That demo key is scoped to one project, vscode-docs, so every example below searches the VS Code documentation set.

Make your first call

Send a POST request to /v1/knowledge/search. Put your questions in the queries array, the project name in project, and your key in the X-API-Key header.

import requests

response = requests.post(
    "https://api.ragionex.com/v1/knowledge/search",
    headers={"X-API-Key": "rgx_knowledge_demo_geXuBZJ5O2GltcG63s1LGijGBwlERGlo"},
    json={
        "queries": ["How do I format code in VS Code?"],
        "project": "vscode-docs",
        "results": 5
    }
)
data = response.json()
for r in data["results"]:
    print(r["answer"], "-", r["source"])

const res = 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({
    queries: ["How do I format code in VS Code?"],
    project: "vscode-docs",
    results: 5
  })
});
const data = await res.json();

curl -X POST https://api.ragionex.com/v1/knowledge/search \
  -H "Content-Type: application/json" \
  -H "X-API-Key: rgx_knowledge_demo_geXuBZJ5O2GltcG63s1LGijGBwlERGlo" \
  -d '{
    "queries": ["How do I format code in VS Code?"],
    "project": "vscode-docs",
    "results": 5
  }'

The request body has three fields:

• queries (required) - an array of 1-5 natural-language questions, each 1-128 characters. One question = a one-element array; add more items to search several angles in a single call.
• project (required) - the documentation set to search. Use vscode-docs with the demo key.
• results (optional) - how many results to return, 1-50. Defaults to 10.

Read the result

You get back the answer as ground truth - the prepared documentation answer, not a generated guess. Each result carries the answer text, the source file it came from, and a stable id you can log.

{
  "success": true,
  "results": [
    {
      "answer": "To format your code in VS Code, use the Format Document command (Shift+Alt+F) or enable Format On Save in settings...",
      "source": "docs/editor/codebasics.md",
      "id": "X1Y2Z3"
    }
  ]
}

Because the answer is prepared content rather than generated text, you can hand it straight to your own AI as trusted context, cite the source, or show it to the user as-is.