Settings
Runtime configuration status for local demos and Splunk-backed operation.
Runtime configuration status for local demos and Splunk-backed operation.
Runtime configuration status for local demos and Splunk-backed operation.
Bring your own Gemini key for live AI reports. Stored only in this browser for the demo.
When you run the resolver, Talos sends this key directly to the Node API route for that request only. A BYOK Gemini key is required for real AI triage reports.
Do not put this key in the SDK. Browser apps should send errors to Talos ingest only; server routes handle AI and Splunk calls.
Configure your Splunk HTTP Event Collector. Sends captured errors directly to your index.
When a real SDK event is captured, Talos routes the error data through your client credentials.
Ensure your Splunk instance allows traffic from this host, and HTTP Event Collector (HEC) is enabled under Settings > Data Inputs.
Configure Model Context Protocol. Connects the AI resolver directly to your local Splunk MCP host.
The Splunk MCP server enables LLMs to query Splunk indexes securely.
When the resolver runs, Talos directs query operations to this browser-configured MCP endpoint. It acts as the primary telemetry gatherer before fallback to cached telemetry.
Pushes triaged incident summaries and fix recommendation codes to Slack or Discord.
When reports are successfully generated, notifications are dispatched to these target browser channels.
Discord webhooks receive complete Markdown formatting including syntax-highlighted code-diff proposals.
import { Talos } from "@mylife-as-miles/talos-sdk";
Talos.init({
projectKey: "demo_project_key",
environment: "production",
release: "v2.4.1",
service: "catalog-service",
ingestUrl: "/api/ingest",
});