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

# AI SRE Instructions

> Add team-specific guidance that steers what Rootly AI SRE checks and how it reports every investigation.

AI SRE instructions capture the investigation practices your team wants Rootly AI SRE to follow across alerts and incidents. Use them for durable guidance about evidence, system conventions, and report quality.

## Configure Instructions

Add guidance that applies to every AI SRE investigation for the current team.

<Steps>
  <Step title="Open AI SRE Settings">
    Open **AI & Agents → AI SRE**, then select **Instructions**.
  </Step>

  <Step title="Write The Guidance">
    Describe the systems to check, evidence to prioritize, or investigation practices to follow. Use direct, testable language.
  </Step>

  <Step title="Save The Instructions">
    Save the editor. The guidance applies to future investigations for everyone on the current team.
  </Step>
</Steps>

Incident Response Owners, Incident Response Admins, and On-Call Admins who have an Incident Response seat can manage AI SRE instructions after your Rootly account team enables AI SRE for the current team. Rootly checks the role and seat on that team; a qualifying membership on another team does not grant access. Contact your Rootly representative to disable AI SRE; the self-service global AI opt-in is not its kill switch.

## What To Include

Useful instructions teach AI SRE something specific about your environment or evidence standards.

<ParamField path="System Conventions">
  Name the service aliases, environment labels, ownership boundaries, or topology details that aren't obvious from provider data.
</ParamField>

<ParamField path="Evidence Priorities">
  Identify the source of truth for a signal. For example: "Use the payment-api error-rate monitor to establish customer impact; don't infer impact from worker retry volume."
</ParamField>

<ParamField path="Investigation Practices">
  State checks that should happen across every investigation. For example: "Compare the alert window with the previous 30 minutes and check deploys to the affected service and its direct dependencies."
</ParamField>

<ParamField path="Reporting Requirements">
  Define what a useful handoff contains. For example: "Separate confirmed facts from open hypotheses and name the missing evidence for every unresolved cause."
</ParamField>

## Instruction Scope And Precedence

Rootly can apply guidance at several scopes.

| Scope                               | Applies To                                            | Behavior                                                  |
| ----------------------------------- | ----------------------------------------------------- | --------------------------------------------------------- |
| **Global instructions**             | Every supported Rootly AI feature on the current team | General team-wide guidance                                |
| **AI SRE instructions**             | Every AI SRE alert and incident investigation         | More specific than global instructions when they conflict |
| **Investigation rule instructions** | Alerts matching one non-paused rule                   | Added to the AI SRE guidance for that alert class         |

Configure global instructions under **AI & Agents → Global → Instructions**. See [AI Settings](/ai/ai-settings) for the team-wide AI configuration surface.

Only the matching non-paused investigation rule with the lowest stored position contributes rule-specific instructions. Sorting the rules table doesn't change that priority, and manual reordering isn't currently available. Keep overlapping rules consistent, or make their conditions mutually exclusive.

<Note>
  Instructions steer where AI SRE looks and how it reports. They don't grant permissions, expand connector access, override Private Agent policy, or prove a cause. The investigation still needs current evidence.
</Note>

## Write Effective Instructions

* Name the provider, service, signal, or environment that matters.
* Ask for a comparison or verification that produces evidence.
* Explain ambiguous naming, such as two provider labels that refer to the same service.
* State how to handle uncertainty and what to include in the report.
* Keep team-wide instructions broadly applicable. Put alert-specific guidance in an investigation rule.

Avoid instructions that:

* Assert a cause before the investigation starts.
* Tell AI SRE to ignore contradictory evidence.
* Prescribe a production change without a current safety check.
* Depend on one incident's timestamp, identifier, or temporary state.
* Repeat information already present in every alert payload.

## Examples

### Evidence Standard

```text theme={null}
Treat the checkout service level objective (SLO) as the source of truth for
customer impact.
Corroborate it with payment-api errors and traces. If the sources disagree,
report the disagreement instead of choosing one silently.
```

### Deployment Correlation

```text theme={null}
For a service regression, check deploys to the affected service and its direct
dependencies during the 30 minutes before the signal changed. A nearby deploy
is a candidate cause until logs, metrics, or traces show the mechanism.
```

### Report Quality

```text theme={null}
Separate confirmed facts, likely causes, and open questions. Every recommended
next step must name the signal that would confirm success.
```

## Related Pages

<CardGroup cols={2}>
  <Card title="Investigation Rules" icon="sliders" href="/ai/ai-sre/investigation-rules">
    Apply guidance to a specific class of alerts and choose when it runs automatically.
  </Card>

  <Card title="Memory" icon="brain" href="/ai/ai-sre/memory">
    Curate evidence-backed knowledge that AI SRE can retrieve when it is relevant.
  </Card>
</CardGroup>

## Frequently Asked Questions

<AccordionGroup>
  <Accordion title="Do instructions replace connectors?" icon="plug">
    No. Instructions tell AI SRE what to check, while an AI connector or Private Agent provider supplies the evidence. If the required source isn't connected or permitted, AI SRE reports the gap.
  </Accordion>

  <Accordion title="Should a runbook go in instructions or memory?" icon="book-open">
    Put a short, broadly applicable investigation practice in instructions. Keep team-specific facts and reusable diagnostic learnings in Memory, and connect the full runbook through a knowledge provider such as Notion, Confluence, or a custom Model Context Protocol (MCP) server.
  </Accordion>

  <Accordion title="What happens when a rule conflicts with AI SRE instructions?" icon="triangle-exclamation">
    Both pieces of AI SRE guidance are supplied to the investigation. Avoid conflicting wording by keeping team-wide instructions general and rules specific to their alert class.
  </Accordion>
</AccordionGroup>