Skip to content
POST/api/v1/generators/rag-auto-eval

Auto-generate RAG eval suite

Synthesizes a RAG-evaluation test suite from a knowledge corpus using a real LLM call. `projectId` and at least one `{id, content}` document are required; `casesPerDocument` defaults to 3 and must be 1–20, `questionTypes` is a subset of factual/inferential/comparison/multi-hop/unanswerable/edge_case/adversarial, and total document content is capped at 5 MB (413). Documents are chunked, then per question type the generator produces question + reference-answer pairs with per-case scorer assertions and thresholds drawn from context-relevance, answer-relevance, context-faithfulness, factual-correctness, reasoning-quality, completeness, multi-document-retrieval, hallucination-detection, refusal-accuracy, edge-case-handling and adversarial-robustness. Returns `testCases`, `totalGenerated`, `documentCount`, `chunkCount`, `questionTypeBreakdown`, `duration`, `generatorModel` (default `gpt-4o-mini`) and `keySource`. Uses the org's BYOK OpenAI key, falling back to the platform key; 503 `NO_LLM_KEY` if neither is configured. Rate-limited 10/min.

Authentication

Send Authorization: Bearer YOUR_API_KEY on every request. Generate API keys at /dashboard/settings/api-keys.

Request body required

Example

{
  "corpus_id": "00000000-0000-0000-0000-000000000000",
  "sample_count": 50
}
Schema
{
  "application/json": {
    "schema": {
      "type": "object",
      "required": [
        "corpus_id"
      ],
      "properties": {
        "corpus_id": {
          "type": "string",
          "format": "uuid"
        },
        "sample_count": {
          "type": "integer",
          "default": 50
        }
      }
    }
  }
}

Response

202 example

{
  "job_id": "00000000-0000-0000-0000-000000000000"
}

All status codes

200OK.
202Generation queued.
400(no description)
401(no description)
403Forbidden — insufficient role for this operation.
413Payload Too Large — CONTENT_TOO_LARGE.
429(no description)
500Internal Server Error — GENERATION_ERROR.
503Service Unavailable — NO_LLM_KEY.

Code samples

cURL

curl -X POST \
  https://evalguard.ai/api/v1/generators/rag-auto-eval \
  -H "Authorization: Bearer $EVALGUARD_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "corpus_id": "00000000-0000-0000-0000-000000000000", "sample_count": 50 }'

TypeScript

// The TypeScript SDK (@evalguard/sdk) exposes TYPED methods — runEval,
// getEval, runSecurityScan, checkFirewall, … — not a generic request().
// For an arbitrary endpoint, call it directly:

const res = await fetch("https://evalguard.ai/api/v1/generators/rag-auto-eval", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.EVALGUARD_API_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "corpus_id": "00000000-0000-0000-0000-000000000000",
    "sample_count": 50
  }),
});
console.log(res.status, await res.json());

Python

# The Python SDK (pip install evalguardai) exposes TYPED methods on
# EvalGuardClient — run_eval, get_eval, … — not a generic request().
# For an arbitrary endpoint, call it directly:

import os
import requests

headers = {"Authorization": f"Bearer {os.environ['EVALGUARD_API_KEY']}"}
headers["Content-Type"] = "application/json"

response = requests.request(
    "POST",
    "https://evalguard.ai/api/v1/generators/rag-auto-eval",
    headers=headers,
    json={
    "corpus_id": "00000000-0000-0000-0000-000000000000",
    "sample_count": 50
},
)
print(response.status_code, response.json())

Go

package main

import (
	"context"
	"fmt"
	"net/http"
	"os"
	"strings"
)

func main() {
	body := strings.NewReader(`{"corpus_id":"00000000-0000-0000-0000-000000000000","sample_count":50}`)
	req, _ := http.NewRequestWithContext(context.Background(), "POST", "https://evalguard.ai/api/v1/generators/rag-auto-eval", body)
	req.Header.Set("Authorization", "Bearer "+os.Getenv("EVALGUARD_API_KEY"))
	req.Header.Set("Content-Type", "application/json")
	resp, err := http.DefaultClient.Do(req)
	if err != nil { panic(err) }
	defer resp.Body.Close()
	fmt.Println(resp.Status)
}

Errors

400401403413429500503

Other Evals endpoints