Developer tools
Read and update Sentry issues from your workflows.
Agents triage new issues, read stack traces and recent events, resolve or archive what is handled, and report on releases. Workflows can start from supported issue, alert, Seer, and comment webhooks.
5,000 free credits to start.
Agent reach
Sentry REST API
The full surface, searched at run time.
Connect with
- API key
- Your own OAuth app
Use the token of a Sentry internal integration, or your own public integration through OAuth. For triggers, point the same integration's webhook at Helios.
Event triggers
Yes · 6 events
Plus manual, schedule, and signed webhook.
What the agent can do
What agents do with Sentry
Helios searches the Sentry REST API at run time. The agent finds the endpoint the task needs and calls it.
- Call Sentry's REST API with the connected integration's permissions.
- Find the newest or most frequent issues in a project, environment, or release.
- Read an issue's events, stack trace, breadcrumbs, and tags.
- Resolve, archive, assign, or delete issues, and change their priority.
- Read an issue's comments and reply to them.
- Create releases and deploys, and manage alerts and monitors.
- Start a run when an issue changes, an error arrives, an alert fires, Seer reaches a stage of a fix, a mobile build is processed, or someone comments.
Examples
Tasks you can hand to a Sentry agent
Investigate a metric monitor alert
When error rates cross a threshold, the agent checks what shipped recently and posts what it found to the on-call channel.
When a metric monitor's alert fires, compare it with recent deploys and post a summary to #oncall.
- sentry·read the alert and its query
- sentry·list releases deployed in the last hour
- sentry·find the issues that grew the most
- slack·post a summary to #oncall
Error spike traced to release 2026.10.1
- Errors rose from 40 to 900 a minute after the 14:05 deploy
- Most come from one new issue in the payments service
- Posted to #oncall with a link to the release
Clean up stale issues
A scheduled agent reviews unresolved issues that have not been seen in weeks and resolves or archives them.
Every Monday, resolve issues not seen in 30 days and archive known noisy ones.
- sentry·list unresolved issues last seen 30+ days ago
- sentry·resolve the stale issues
- sentry·archive known noisy issues until they escalate
- slack·post what changed to #eng
Weekly Sentry cleanup
- Resolved 37 issues not seen in 30 days
- Archived 5 browser extension errors
Event triggers
Start a run on Sentry events
Attach any of these events to a workflow and filter on the fields that matter. Manual, schedule, and signed webhook triggers work too.
| Event | What it means |
|---|---|
issue.created | A new issue appears in a project - the moment to triage it and file a ticket. |
issue_alert.triggered | An issue alert sends a notification to Helios. Filter on the project, environment, or alert name. |
issue.unresolved | A resolved issue regresses or an archived one escalates, so an agent can find what changed. |
activity_alert.triggered | An alert fires when an issue resolves, regresses, or reaches a Seer stage, so an agent can follow up. |
seer.root_cause_completed | Seer finds the root cause of an issue, so an agent can file a ticket with it. |
comment.created | Someone comments on an issue, so an agent can pick up the request. |
Works with
Sentry works with the rest of your stack
One agent, one run, several services. It reads from one tool and writes to the next in the same sandbox.
FAQ
Questions, answered
Put an agent on Sentry.
Connect it once and describe the task in chat. Then reuse the prompt as a workflow on a schedule or a webhook.