Skip to content
Draft
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
53 changes: 53 additions & 0 deletions docs/ai/agent-skills/claude.mdx
Original file line number Diff line number Diff line change
@@ -0,0 +1,53 @@
---
title: Claude Agent Setup
sidebar_order: 10
description: "Set up the Claude Agent integration to send Sentry issues directly to Claude for automated root cause analysis and fix generation."
keywords:
- Claude
- Claude Agent
- agent skills
- AI coding assistant
- integration
---

You can trigger Claude agents from your [Seer Autofix](/product/ai-in-sentry/seer/autofix/) tab. The agent is provided with Seer's root cause analysis output and issue context, so it can understand the problem and generate a fix — including opening a branch with the changes.

## Install and Configure

<Alert>

Sentry Owner, Manager, or Admin permissions are required to install this integration. You must also have an existing Claude workspace and environment set up before connecting to Sentry.

</Alert>

1. Go to [platform.claude.com/dashboard](https://platform.claude.com/dashboard) and create an API key for your workspace.

{/* TODO: screenshot - Claude dashboard API key creation */}

2. In Sentry, navigate to **Settings > Integrations** and search for **Claude Agent**.

{/* TODO: screenshot - Integrations page with Claude Agent in search results */}

3. Enter your API key. If your workspace isn't "default" or you want to use a specific environment, update those fields here, then save.

{/* TODO: screenshot - Claude Agent integration config form (empty) */}

{/* TODO: screenshot - Claude Agent integration config form (filled) */}

{/* TODO: screenshot - Claude Agent integration config form (saved/confirmed) */}

## Using the Integration

Once installed, you can send any Seer root cause analysis to a Claude agent.

1. Go to a Sentry issue and click **Start Root Cause Analysis**.

{/* TODO: screenshot - Sentry issue page with "Start Root Cause Analysis" button */}

{/* TODO: screenshot - Root cause analysis running/completed */}

2. Once root cause analysis completes, open the dropdown and select **Send to Claude Agent**.

{/* TODO: screenshot - dropdown with "Send to Claude Agent" option */}

3. A card will appear in the drawer for the active session. While the session is running, you can follow along in the Claude Console. Once complete, a button linking to the created branch will appear in the card.
51 changes: 51 additions & 0 deletions docs/ai/agent-skills/copilot.mdx
Original file line number Diff line number Diff line change
@@ -0,0 +1,51 @@
---
title: GitHub Copilot Agent Setup
sidebar_order: 20
description: "Set up the GitHub Copilot integration to send Sentry issues directly to Copilot agents for automated root cause analysis and fix generation."
keywords:
- GitHub Copilot
- Copilot
- agent skills
- AI coding assistant
- integration
---

You can trigger GitHub Copilot agents from your [Seer Autofix](/product/ai-in-sentry/seer/autofix/) tab. The agent is provided with Seer's root cause analysis output and issue context, so it can understand the problem and generate a fix, including opening a pull request in GitHub.

## Install and Configure

<Alert>

Sentry Owner, Manager, or Admin permissions are required to install this integration.

</Alert>

1. In Sentry, navigate to **Settings > Integrations** and search for **GitHub Copilot**.

2. Click **Install**.

<br />

<Arcade src="https://demo.arcade.software/d5AS2f3OiW6sIdDp9crY?embed&show_copy_link=true" width="65%" />

## First-Time Authorization

The first time you use the integration, you'll need to authorize Copilot from within a Sentry issue.

1. Go to a Sentry issue and click **Start Root Cause Analysis**.

2. Once root cause analysis completes, click **Set Up GitHub Copilot** in the Seer panel.

3. Complete the authorization and redirect flow to connect your GitHub account.

## Using the Integration

Once authorized, you can send any Seer root cause analysis to a Copilot agent.

1. On any Seer root cause analysis, open the dropdown and select **Send to GitHub Copilot**.


2. A link to the Copilot agent run and any resulting pull request will appear in the Sentry UI. Click it to follow Copilot's progress and review the changes.

<br />
<Arcade src="https://demo.arcade.software/VD4hY2rxo3i42u3fXb3j?embed&show_copy_link=true" width="65%" />
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
File renamed without changes.
44 changes: 28 additions & 16 deletions src/components/arcade.tsx
Original file line number Diff line number Diff line change
@@ -1,30 +1,42 @@
type ArcadeProps = {
src: string;
/** Constrains the embed width (for example `640px` or `80%`). Defaults to full width of the content column. */
width?: string;
};
export function Arcade({src}: ArcadeProps) {

export function Arcade({src, width}: ArcadeProps) {
return (
<div
style={{
height: '0px',
paddingBottom: 'calc(56.8359% + 41px)',
position: 'relative',
width: '100%',
maxWidth: '100%',
width: width ?? '100%',
}}
>
<iframe
src={src}
loading="lazy"
allowFullScreen
allow="fullscreen;"
{/* Inner wrapper: % padding is relative to outer width so aspect ratio matches narrow embeds. */}
<div
style={{
colorScheme: 'light',
height: '100%',
left: '0px',
position: 'absolute',
top: '0px',
height: '0px',
paddingBottom: 'calc(56.8359% + 41px)',
position: 'relative',
width: '100%',
}}
/>
>
<iframe
src={src}
loading="lazy"
allowFullScreen
allow="fullscreen;"
style={{
border: 'none',
colorScheme: 'light',
height: '100%',
left: '0px',
position: 'absolute',
top: '0px',
width: '100%',
}}
/>
</div>
</div>
);
}
Loading