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

# On Call

> Connect Indent to your incident response tooling and databases so it can help you RCA and fix production issues fast.

When something breaks in production, Indent can pull together the context your team needs — logs, traces, error reports, and database queries — to root-cause the issue and get to a fix faster.

**Best with:** [Slack](/slack), [Datadog](/datadog), [Sentry](/sentry), and [Databases](/databases).

## Connecting your tooling

From the [Integrations](https://app.indent.com/settings/services) page, connect the services Indent should have access to during an investigation:

* **Slack** — lets your team kick off sessions by mentioning `@indent-zero` directly in an incident thread or DM.
* **Datadog** — pull metrics, traces, and logs to diagnose production issues.
* **Sentry** — pull error reports and stack traces to debug faster.
* **Databases** — connect PostgreSQL, BigQuery, Snowflake, or other databases from the [Database Access](https://app.indent.com/settings/databases) page so Indent can query your data directly during an investigation.

## Starting an investigation

You can start an investigation from chat or straight from Slack. Paste a Sentry error link, describe what's happening, or mention `@indent-zero` in an incident channel — Indent picks it up and starts working.

From there, Indent pulls in context from your connected services: querying Datadog for recent metrics, checking Sentry for related errors, and running database queries to trace the issue through your data. It asks clarifying questions when it needs to — what the user was doing, whether there were recent deploys, what the expected behavior was — then works through the evidence to identify a root cause.

## From root cause to fix

Once Indent has an RCA, the same session can continue into a code fix. Because Indent already has the full investigation context — the error, the traces, the data it queried — it can open a pull request with the fix without you having to re-explain the problem. See [Code](/code) for how that flow works.