> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cascadeflow.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Observability And Privacy

> How to use traces, summaries, and telemetry while keeping the rollout privacy-aware.

Use this page when you need runtime visibility but also need to think clearly about data exposure and auditability.

## What cascadeflow Produces

| Surface                     | Why it matters                                                    |
| --------------------------- | ----------------------------------------------------------------- |
| Session summaries           | Quick operational view of cost, steps, tools, latency, and budget |
| Decision traces             | Explain why a step was allowed, switched, denied, or stopped      |
| Framework-specific metadata | Makes traces available in the runtime systems teams already use   |

## Why This Matters

* Transparency is part of the value proposition, not a side effect.
* Regulated or high-stakes workflows need attributable runtime decisions.
* Teams need traces for tuning, but they also need to be deliberate about where that data goes.

## Start Here

* [/harness/decision-trace](/harness/decision-trace)
* [/harness/overview](/harness/overview)
* [harness\_telemetry\_privacy.md](https://github.com/lemony-ai/cascadeflow/blob/main/docs/guides/harness_telemetry_privacy.md)

## Operational Guidance

* Start in `observe` to understand the trace volume and shape.
* Route telemetry only to systems that match your privacy requirements.
* Use structured traces to support debugging, audits, and rollout reviews.