Skip to main content
Each integration gives the agent access to new data and actions. All integrations are configured in Settings → Connections. For a full list of what the agent can do with each tool, see the tools reference.

GitHub

Connected automatically during setup. No additional configuration needed.

Slack

The primary way most teams interact with Macroscope. DM Macroscope or tag @Macroscope in any channel to query the agent. Macroscope also streams real-time code activity into Slack channels via /macroscope commands. Use /macroscope subscriptions to view a channel’s subscriptions, or /macroscope help for the full command list.
  • /macroscope subscribe <repo>: all activity from a repo
  • /macroscope subscribe-commits <repo>: commit summaries (includes linked Jira and Linear tickets)
  • /macroscope subscribe-prs <repo>: PR activity, with live emoji reactions as PRs are ✅ approved, 💬 commented, or 🚀 merged
  • /macroscope subscribe-build-failures <repo>: build failures
Add --paths= to any command to filter to specific parts of a repo (e.g. --paths=/frontend).
A Slack admin can connect Slack in Settings → Connections → Slack.

Linear

The agent can look up issues, track project progress, and create new issues. When you connect Linear, you choose which teams Macroscope can access.
1

Go to Settings → Connections

Scroll down to the Linear section.
2

Sign in to Linear

Click Connect Linear and sign in with your Linear account.
3

Choose Team Access

Linear will prompt you to choose which teams Macroscope can access. Select all public teams or only select teams.Linear permissions selection exampleClick Authorize to complete.

Jira

The agent can look up tickets, track project progress, and create new issues. Macroscope inherits the permissions of the Jira account that authorizes the integration.
We recommend creating a dedicated Jira user for Macroscope, scoped to the specific Projects you want Macroscope to access. For Google Workspace teams, create an email alias like linda+macroscope@yourcompany.com and use it for the Jira account.
1

Go to Settings → Connections

Scroll down to the Jira section.
2

Sign in to Jira

Make sure you are logged out of any personal Jira accounts first. Click Connect Jira and sign in with the account you want to authorize for Macroscope.
In some cases, a Jira site admin may need to manually authorize the connection:
  1. Go to Atlassian Administration → Apps → [your-site].atlassian.net → Connected apps
  2. Find Macroscope and click Authorize/Allow for the site.
3

Review Project List

You’ll see a list of Projects that Macroscope has been granted access to. These will be accessible to anyone on your team with Macroscope access.To modify the list, update the permissions of the authorizing Jira account and re-authorize.Click Confirm Access to complete.

LaunchDarkly

The agent can list feature flags, projects, and environments, and check flag status and targeting rules.
1

Go to Settings → Connections

Scroll down to the LaunchDarkly section.
2

Generate an Access Token

In LaunchDarkly, go to Settings → Authorization → Create Token. Assign the Reader role.
3

Paste Token

Copy the token and paste it into the Macroscope connection modal. Click Connect.

PostHog

The agent can query product analytics events, analyze trends, and inspect feature flags.
1

Go to Settings → Connections

Scroll down to the PostHog section.
2

Create a Personal API Key

In PostHog, go to Settings → Personal API Keys → Create. Copy the key.
Use an organization-scoped API key for flexibility across multiple projects.
3

Connect to Macroscope

Select your region (US or EU), paste your API key, and click Next.
4

Select a Project

Macroscope will auto-fetch your projects. Select a default project and click Connect.

Sentry (beta)

The agent can search error issues across projects, view issue details and event history, and investigate crashes and exceptions.
1

Go to Settings → Connections

Scroll down to the Sentry section.
2

Select Your Instance

Choose your Sentry region: sentry.io, us.sentry.io, de.sentry.io, or enter a custom URL for self-hosted instances.
3

Generate an Auth Token

In Sentry, go to Settings → Auth Tokens and create a token with the following scopes: event:read, project:read, org:read.
4

Connect

Paste the auth token, enter your organization slug, and click Connect. Macroscope will validate your credentials automatically.

BigQuery

The agent can browse datasets, inspect table schemas, and run read-only SQL queries. Answers include a methodology summary so you can verify the analysis.
1

Go to Settings → Connections

Scroll down to the BigQuery section.
2

Create a GCP Service Account

In Google Cloud Console, create a service account with BigQuery Data Viewer and BigQuery Job User roles. Generate a JSON key file.
3

Connect to Macroscope

Enter a connection name, your GCP Project ID, and upload the service account JSON key (or paste it directly). Optionally add prompt instructions to give the agent context about your data.Click Create to complete.

GCP Cloud Logging

The agent can search and filter application logs across connected GCP projects.
1

Go to Settings → Connections

Scroll down to the GCP Cloud Logging section.
2

Create a GCP Service Account

In Google Cloud Console, create a service account with Logs Viewer role. Generate a JSON key file.
3

Connect to Macroscope

Enter a connection name, your GCP Project ID, and upload the service account JSON key (or paste it directly). Optionally add prompt instructions to give the agent context about your logs.Click Create to complete.

Amplitude (beta)

The agent can run event segmentation queries, analyze conversion funnels and retention, and inspect individual user event streams.
1

Go to Settings → Connections

Scroll down to the Amplitude section.
2

Get Your API Credentials

In Amplitude, find your API Key and Secret Key in your project settings.
3

Connect to Macroscope

Enter a connection name, select your region (US or EU), paste your API Key and Secret Key, and optionally add prompt instructions.Click Create to complete.

MCP Connectors (beta)

Connect any MCP-compatible server to extend the agent with custom tools (e.g. Datadog, PagerDuty, or your own internal APIs). Admin only.
1

Go to Settings → Connections

Scroll down to the MCP Connectors section.
2

Enter Server URL

Paste the MCP server URL and click Discover. Macroscope will auto-detect the server’s authentication method.
3

Configure Authentication

Depending on the server, you may need to provide OAuth credentials (Client ID and Secret), custom headers, or no authentication.
4

Select Tools

Review the tools the server provides. Toggle off any tools you don’t want the agent to use. Click Connect.