Documentation Index
Fetch the complete documentation index at: https://docs.trodo.ai/docs/llms.txt
Use this file to discover all available pages before exploring further.
Keep your existing OTel pipeline (Honeycomb, Datadog, Grafana Tempo, Jaeger) and add Trodo alongside. The Trodo SDK doesn’t own the OTel tracer — it registers its own exporter. Any other exporter you’ve set up keeps working.
import { NodeTracerProvider } from '@opentelemetry/sdk-trace-node';
import { BatchSpanProcessor } from '@opentelemetry/sdk-trace-base';
import { OTLPTraceExporter } from '@opentelemetry/exporter-trace-otlp-http';
import trodo from 'trodo-node';
// 1. Your existing OTel setup — unchanged.
const provider = new NodeTracerProvider();
provider.addSpanProcessor(
new BatchSpanProcessor(
new OTLPTraceExporter({ url: 'https://api.honeycomb.io/v1/traces' }),
),
);
provider.register();
// 2. Trodo attaches its own exporter to the same provider.
trodo.init({ siteId: process.env.TRODO_SITE_ID });
// 3. Instrument as normal. Every span goes to both exporters.
await trodo.wrapAgent('support-bot', async () => {
// ...
});
from opentelemetry import trace
from opentelemetry.sdk.trace import TracerProvider
from opentelemetry.sdk.trace.export import BatchSpanProcessor
from opentelemetry.exporter.otlp.proto.http.trace_exporter import OTLPSpanExporter
import trodo, os
# 1. Your existing OTel setup — unchanged.
provider = TracerProvider()
provider.add_span_processor(BatchSpanProcessor(OTLPSpanExporter(endpoint="https://api.honeycomb.io/v1/traces")))
trace.set_tracer_provider(provider)
# 2. Trodo attaches its own exporter to the same provider.
trodo.init(site_id=os.environ["TRODO_SITE_ID"])
# 3. Instrument as normal.
with trodo.wrap_agent("support-bot") as run:
...
What each backend sees
- Your OTel collector — standard spans with
gen_ai.* semantic-convention attributes. Your existing dashboards keep working.
- Trodo — the same spans, grouped into runs, with the Trodo-specific rollups (tokens, cost, feedback, clusters).
Both use the same trace_id / span_id, so you can cross-reference a run in Trodo against a trace in your OTel backend.
Disable Trodo’s built-in exporter
If you’d rather pipe everything through your existing OTLP collector and let it forward to Trodo, set exporter: 'none':
trodo.init({ siteId: process.env.TRODO_SITE_ID, exporter: 'none' });
https://sdkapi.trodo.ai/otlp/v1/traces with header X-Trodo-Site-Id: <your-site-id>.
See also
