How do AI agents use Sentry through Open Connector?

Your user connects Sentry once with one of its cataloged authentication methods. Open Connector stores the credential in an encrypted vault and exposes Sentry tools to your agent over MCP or a typed API, with credentials injected server-side on each call.

Is this a Sentry MCP server?

Yes. Open Connector can serve Sentry as a named MCP server with a scoped allowlist and a per-user connection URL, so any MCP client can call Sentry actions with credentials injected server-side.

Where do Sentry credentials live?

In your own infrastructure. Open Connector keeps credentials in its own vault and injects them at call time, so they never leave your environment.

Sentry integration

Sentry
for AI agents.

Sentry integration for AI agents with secure authentication and server-side credential injection. Open Connector runs the OAuth, seals the token in an encrypted vault, and serves Sentry tools to your agent over MCP or a typed API — credentials injected server-side, every call audited, nothing leaving your infrastructure. Open source (AGPL-3.0) and self-hostable.

View source on GitHub All integrations →

What your agents can do

Real Sentry actions, managed and audited.

Your user connects Sentry once; your agent can then manage projects, issues, releases, and integrations — scoped to the OAuth permissions you grant and the tool allowlist you configure. Every action is least-privilege and written to a tamper-evident audit trail.

01Connect
Your user grants Sentry access once (OAuth) — the token lands in the vault.
02Call
Your agent calls a tool over MCP or the typed API; Open Connector injects the credential server-side.
03Audit
Every routed call appends a hash-chained audit record — nothing leaves your infra.

Tools & triggers

Supported Sentry tools.

Browse 17 tools in the Sentry catalog. Each action uses your configured authentication and tool allowlist.

Showing 17 tools.

Bulk Mutate A List Of Issues: Bulk mutate various attributes on issues. The list of issues to modify is given through the `id` query parameter. It is repeated for each issue that should be modified. - For non-status updates, the `id` query parameter is required. - For status updates, the `id` query parameter may be omitted for a batch "update all" query. - An optional `status` query parameter may be used to restrict mutations to only events with the given status. The following attributes can be modified and are supplied as JSON object in the body: If any IDs are out of scope this operation will succeed without any data mutation.; Events
Create Or Update An External Issue: Create or update an external issue from an integration platform integration.; Integration
Delete An External Issue: Delete an external issue.; Integration
Disable Spike Protection: Disables Spike Protection feature for some of the projects within the organization.; Projects
Enable Spike Protection: Enables Spike Protection feature for some of the projects within the organization.; Projects
List An Organization'S Integration Platform Installations: Return a list of integration platform installations for a given organization.; Integration
List A Project'S Issues: **Deprecated**: This endpoint has been replaced with the [Organization Issues endpoint](/api/events/list-an-organizations-issues/) which supports filtering on project and additional functionality. Return a list of issues (groups) bound to a project. All parameters are supplied as query string parameters. A default query of ``is:unresolved`` is applied. To return results with other statuses send an new query value (i.e. ``?query=`` for all results). The ``statsPeriod`` parameter can be used to select the timeline stats which should be present. Possible values are: ``""`` (disable),``"24h"`` (default), ``"14d"`` User feedback items from the [User Feedback Widget](https://docs.sentry.io/product/user-feedback/#user-feedback-widget) are built off the issue platform, so to return a list of user feedback items for a specific project, filter for `issue.category:feedback`.; Events
List A Project'S Service Hooks: Return a list of service hooks bound to a project.; Projects
List A Project'S User Feedback: Return a list of user feedback items within this project. *This list does not include submissions from the [User Feedback Widget](https://docs.sentry.io/product/user-feedback/#user-feedback-widget). This is because it is based on an older format called User Reports - read more [here](https://develop.sentry.dev/application/feedback-architecture/#user-reports). To return a list of user feedback items from the widget, please use the [issue API](https://docs.sentry.io/api/events/list-a-projects-issues/) with the filter `issue.category:feedback`.*; Projects
List A Tag'S Values: Return a list of values associated with this key. The `query` parameter can be used to to perform a "contains" match on values. When [paginated](/api/pagination) can return at most 1000 values.; Projects
List A Tag'S Values Related To An Issue: Returns details for given tag key related to an issue. When [paginated](/api/pagination) can return at most 1000 values.; Events
Register A New Service Hook: Register a new service hook on a project. Events include: - event.alert: An alert is generated for an event (via rules). - event.created: A new event has been processed. This endpoint requires the 'servicehooks' feature to be enabled for your project.; Projects
Remove A Service Hook: Remove a service hook.; Projects
Retrieve A Service Hook: Return a service hook bound to a project.; Projects
Retrieve Files Changed In A Release'S Commits: Retrieve files changed in a release's commits; Releases
Submit User Feedback: *This endpoint is DEPRECATED. We document it here for older SDKs and users who are still migrating to the [User Feedback Widget](https://docs.sentry.io/product/user-feedback/#user-feedback-widget) or [API](https://docs.sentry.io/platforms/javascript/user-feedback/#user-feedback-api)(multi-platform). If you are a new user, do not use this endpoint - unless you don't have a JS frontend, and your platform's SDK does not offer a feedback API.* Feedback must be received by the server no more than 30 minutes after the event was saved. Additionally, within 5 minutes of submitting feedback it may also be overwritten. This is useful in situations where you may need to retry sending a request due to network failures. If feedback is rejected due to a mutability threshold, a 409 status code will be returned. Note: Feedback may be submitted with DSN authentication (see auth documentation).; Projects
Update A Service Hook: Update a service hook.; Projects

FAQ

Sentry integration, answered

How do AI agents use Sentry through Open Connector?: Your user connects Sentry once with one of its cataloged authentication methods. Open Connector stores the credential in an encrypted vault and exposes Sentry tools to your agent over MCP or a typed API, with credentials injected server-side on each call.
Is this a Sentry MCP server?: Yes. Open Connector can serve Sentry as a named MCP server with a scoped allowlist and a per-user connection URL, so any MCP client can call Sentry actions with credentials injected server-side.
Where do Sentry credentials live?: In your own infrastructure. Open Connector keeps credentials in its own vault and injects them at call time, so they never leave your environment.

Give your agents Sentry — keep the keys.

Open source, self-hostable, with Sentry credentials that never leave your infrastructure. Run it from source today.

View source on GitHub How the MCP gateway works →

Sentryfor AI agents.

Real Sentry actions, managed and audited.

Supported Sentry tools.

Sentry integration, answered

Give your agents Sentry — keep the keys.

Sentry
for AI agents.