Platform Getting Started

Connect your agents to the Traccia Platform for full observability.

OTLP Ingestion

How telemetry reaches the Traccia platform — HTTP, gRPC, auth, and which clients use which transport.

Dashboard

Navigate the Traccia dashboard and understand your metrics.

Governance Hub

Org-wide registry, human review queues, incident records, audit bundles, and admin activity logs on top of traces and policies.

Guardrail Posture

See guardrail coverage across traces and agents: what fired, what was absent, and how to tune missing-category noise—without conflating posture with policy enforcement.

Policies

Create guardrails to govern agent behavior.

Prompts

Version and ship the instructions that drive your agents, without redeploying for every wording change.

Prompt Playground

Compare prompt wording and models side by side, with latency, tokens, and cost, before you promote a version.

Datasets

Curate test cases for your prompts: CSV, manual rows, production traces, optional JSON Schema, and graded runs from Prompt Playground.

Scorers

Grade prompt outputs on a dataset: built-in checks, LLM-as-judge, and restricted custom code. Attach scorers in Prompt Playground and save experiments.

Experiments

Save graded dataset runs as immutable evidence, compare baseline vs candidate, then attach the better run when you promote.

Teams

Manage team access and collaboration.

Costs & Pricing

How Traccia tracks, recomputes, and attributes LLM costs across your agents — and how to configure pricing for your organization.

Traces

Explore and analyze traces in the Traccia platform.