Observability

What telemetry Agent Stream produces for every call, what the dashboard shows, and how to export traces, metrics, and audit logs to your own stack.

Agent Stream is the source of truth for AI usage in your organisation, with no external time-series database required to see it, and it still exports to the tools you already run. Trust-fabric integration →

What gets recorded

Signal	What it shows
Usage telemetry	Multi-resolution spend, token, and request time-series, from 30-second to yearly buckets, per surface, stage, member, and team.
Operational metrics	Connection volume, error rate, data throughput, and latency percentiles, at p50, p95, and p99, for fleet health.
Rejection taxonomy	Every guardrail block classified into a stable set of categories, such as hate, toxicity, prompt injection, jailbreak, PII, secrets, and policy violation, with Uncategorized as the fallback, broken down per team, member, surface, and stage.

What the dashboard shows

A real-time management dashboard shows cost, tokens, latency, and throughput, with per-surface breakdowns, and each surface has its own monitoring tab scoped to that surface alone. This is the fastest way to confirm traffic is flowing normally, spot which surface is driving cost, or see a rejection spike as it happens.

The fleet-wide management dashboard, showing active surfaces, latency, total tokens, total cost, and cache metrics alongside cost, token, and traffic charts broken out by surface

What comes back in response headers

A buffered response carries x-agent-stream-* headers: trace ID, provider, model, tokens, cached-tokens, cost, cache hit or miss, and ingress skin. A client can read cost and usage figures directly from the response headers without parsing the body. A streamed response carries the request-time subset of the same headers.

A captured API response showing the x-agent-stream-* headers: trace ID, provider, model, input, output, and total tokens, cost, and a cache hit with the cost it saved

Exporting to your own stack

Nothing here is locked to Agent Stream’s own dashboard. Each export channel carries a different signal:

Channel	What it carries
OpenTelemetry (OTLP)	Traces, metrics, and audit logs such as policy decisions and governance records, to Datadog, Grafana, Jaeger, or any OTLP-compatible collector.
Prometheus scrape endpoint	Metrics.
Webhook integration	Events.

Governance records reach OTLP only when the surface’s OpenTelemetry exporter switch is on, and Enable OpenTelemetry and Enable Log Export are both on in the OpenTelemetry tab of the Metrics Configuration page.

Every request

→

Traces, metrics, audit logs, events

→

OpenTelemetry (traces, metrics, audit logs) · Prometheus (metrics) · Webhook (events)

Why this matters

Usage telemetry and rejection counts carry the same identity context the rest of the platform uses: surface, stage, member, and team. A cost spike or a rejection spike can therefore be attributed to a specific surface or caller on the dashboard, rather than appearing as an undifferentiated failure.

Cost and usage governance: The budgets and alerts built on the same usage telemetry.
Guardrails: The safety layers behind the rejection taxonomy.
Governance records: The signed, per-interaction transcript that complements this operational telemetry.
Teams and attribution: How telemetry rolls up to members and teams.
LLM drift detection: The dedicated dimension charts that sit alongside this operational telemetry.

Was this page helpful?

Glad to hear it! Please tell us how we can improve more.

Sorry to hear that. Please tell us how we can improve.

Thank you for sharing your feedback so we can improve your experience.