Skip to main content

​
Integration Issues

Ensure @Edgedive is invited to the channel and the Slack integration is connected in Connections. Edgedive can only monitor and respond in channels where it has been explicitly invited.
Check that your alerting source (Sentry, Datadog) is connected in Connections and the correct projects/services are selected. For issue tracker triggers (Linear), ensure the issue is assigned to Edgedive or Edgedive is @tagged in a comment.
Go to Connections → GitHub → Manage Repositories and ensure the repo is selected. Some organizations require admin approval for third-party apps.
Verify your cloud log provider credentials have sufficient read permissions. Check that the correct region, log group, or workspace is configured in Agent → Logging Access Providers. See Connections for setup details.
Verify the Sentry organization is selected in Connections → Sentry and your token has read access to the relevant projects.
Double-check that API keys or credentials have the required read permissions. For Grafana, ensure the API key has at least Viewer access. For Datadog, ensure both API and application keys are provided.

​
Investigation Issues

Normal investigation time is under 5 minutes. Complex issues with large codebases or many log sources may take longer. Check the Edgedive dashboard for status. If investigations consistently take too long, try narrowing the scope of connected log sources.
The more integrations connected, the richer the investigation. Ensure relevant observability platforms and cloud log providers are connected. Edgedive produces the best results when it can cross-reference multiple signal sources.
Edgedive improves over time as it processes more alerts. Ensure all relevant data sources are connected. You can ask follow-up questions in the Slack thread to refine the analysis—Edgedive maintains full context throughout the conversation.
Check that GitHub is connected with write access. Common causes: branch protection rules, signed commit requirements, or permissions issues. Try reconnecting GitHub in Connections.
Edgedive runs your existing test suite via configured scripts. Review the test output in the PR and request changes if needed.

​
FAQ

Edgedive is an AI support engineer that automatically investigates on-call engineering alerts. It cross-references logs, metrics, traces, and the codebase to determine root cause, assess impact, and suggest fixes.
General-purpose coding agents require manual log retrieval, custom orchestration, and observability integration to debug production issues.Edgedive is already integrated with your observability stack. It automatically collects production data, builds structured investigations, maintains historical intelligence, and provides secure tenant isolation.Edgedive investigates. Coding agents implement. They are complementary—Edgedive provides the production-grounded context, and coding agents generate the code.
Yes. Edgedive builds on-call intelligence from past alerts and investigations, enabling pattern recognition and faster root cause detection. The more issues it processes, the more accurate investigations become.
Yes. Edgedive functions as a unified interface across logs, metrics, traces, alerts, and code context. Use @Edgedive in Slack to ask natural language questions about your production systems—no need to switch between dashboards.
Yes. After an investigation, you can approve a fix and Edgedive will create a branch, implement the changes, run tests, and open a PR. You can also hand off to an external coding agent with full investigation context.
No. Code is read during analysis but not permanently stored. All data is encrypted and logically isolated per customer. Edgedive is SOC2 Type II and HIPAA compliant.
Edgedive integrates with Sentry, Datadog, Grafana, Prometheus, Loki, SumoLogic, Render, AWS CloudWatch, Azure Monitor, Google Cloud Logging, Slack, Linear, Asana, and GitHub. Jira and ClickUp integrations are coming soon. New integrations can be built in days.
Integration takes minutes. Full setup can be completed in approximately 15 minutes. White-glove onboarding is available.
  • Engineering teams with on-call rotations
  • Teams with heavy alert loads
  • Organizations wanting automated root cause analysis
  • Teams with fragmented logging and observability setups
  • Companies where uptime and reliability are critical (fintech, payments, healthcare, etc.)
Currently GitHub only. Other providers are on the roadmap.
Go to the investigation in the Edgedive dashboard and click Cancel.

​
Contact Support

Email us at team@edgedive.com