Skip to content
GET/api/v1/embeddings/cluster

Cluster embeddings

Returns a 2-D projection of recent agent-step embeddings for scatter visualisation — not a clustering result. `?projectId` is required (400). Optional `since` (ISO, default 7 days ago), `limit` (1-2000, default 500) and `stepType` (llm|tool|retrieval|decision|error). Reads `agent_steps` (inner-joined to `agent_traces` for the project pin) where `embedding_status='completed'` and `embedding` is non-null, newest first. Each point uses the worker-precomputed UMAP coordinate when `embedding_2d` is present (rescaled into [-1,1] on both axes), and falls back to a server-side PCA over the raw embedding otherwise; `method` on each point says which. Returns `{points:[{id, traceId, name, type, isError, createdAt, x, y, method}], summary:{total, byType, errorCount, windowSince, dim, umapPoints, pcaPoints, primaryMethod}}`. Reads through a service-role client, so the projectId pin plus the wrapper's membership check is the tenant boundary. 60 req/min.

Authentication

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

Response

200 example

{
  "success": true
}

All status codes

200Clusters.
400(no description)
401(no description)
403Forbidden — the credential is valid but lacks the API-key scope, member role, or plan entitlement this operation requires.
429(no description)
500Internal Server Error — DB_ERROR.

Code samples

cURL

curl -X GET \
  https://evalguard.ai/api/v1/embeddings/cluster \
  -H "Authorization: Bearer $EVALGUARD_API_KEY"

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/embeddings/cluster", {
  method: "GET",
  headers: { Authorization: `Bearer ${process.env.EVALGUARD_API_KEY}` },
});
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']}"}

response = requests.request("GET", "https://evalguard.ai/api/v1/embeddings/cluster", headers=headers)
print(response.status_code, response.json())

Go

package main

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

func main() {
	req, _ := http.NewRequestWithContext(context.Background(), "GET", "https://evalguard.ai/api/v1/embeddings/cluster", nil)
	req.Header.Set("Authorization", "Bearer "+os.Getenv("EVALGUARD_API_KEY"))
	resp, err := http.DefaultClient.Do(req)
	if err != nil { panic(err) }
	defer resp.Body.Close()
	fmt.Println(resp.Status)
}

Errors

400401403429500

Other Evals endpoints