Govern AI agents on Sentry
Connect Sentry error tracking to governed agent workflows. When Sentry detects errors, agents can auto-triage and respond — with full governance controls ensuring safe remediation.
What this pack does
- Error-triggered agent workflows with governance
- Severity-based policy routing
- Governed auto-triage pipelines
- Error context forwarded to agent decision-making
Use cases
Auto-triage recurring Sentry errors with governed agents
Route critical errors to approval-gated remediation
Log agent responses alongside Sentry issues
Quick setup
- 1Install the Sentry pack: cordumctl pack install sentry
- 2Configure Sentry webhook integration
- 3Map error severity levels to governance tiers
- 4Enable the pack and configure error routing
Frequently asked questions
How do Sentry errors trigger a governed agent workflow?
A Sentry webhook forwards error events to Cordum, which starts an agent workflow with the error context attached. The agent's response actions are evaluated by the Safety Kernel before they run, so auto-triage stays within policy.
Can agents resolve or ignore Sentry issues automatically?
Listing and reading issues, events, and releases are read actions allowed by default, while updating, resolving, ignoring, or muting an issue is a write action governed by policy. That keeps an agent from changing issue state without an Allow or Require Approval decision.
How do I make sure only critical errors reach automated remediation?
Map Sentry severity levels to governance tiers so that critical errors route to approval-gated remediation while routine, recurring errors can be auto-triaged within constraints. Severity-based routing decides which path each error follows.
What context does the agent receive, and what is logged?
The error context from Sentry is forwarded into the agent's decision-making so its response is grounded in the actual issue. Each action and its governance decision is then recorded in the audit trail alongside the originating Sentry issue.
Ready to govern Sentry?
Other integrations
Approval notifications and agent alerts in Slack channels.
Govern AI agent actions on GitHub repositories.
Govern AI agent read and write actions against the AWS API.
Governance for AI agents managing Jira workflows.
Govern AI agents responding to Kubernetes incidents.
Feed Datadog alerts into governed agent workflows.