How to Connect Sentry to an AI Agent: Step-by-Step Setup
· DoubleX
Quick answer
To connect Sentry to an AI agent, open the Connectors screen in DoubleX Agentes, pick Sentry, and authorize access with your Sentry account. Then choose who the connection belongs to (you, a team or the company), limit which actions the agent can run, and confirm it works with a read-only request.
Sentry already holds the errors, releases and performance data your team needs to debug faster. The hard part is getting an AI agent to read it safely, without handing over more access than necessary. This guide covers only the connection: setup, scope, permissions and a first test.
What can an AI agent do with Sentry connected?#
With Sentry connected, an agent can investigate errors, issues, releases and performance data on your behalf. Instead of opening dashboards, you ask a question in plain language and the agent uses Sentry's tools to look it up.
Typical uses once the connection works:
| Task | How the agent helps | Where the data comes from |
|---|---|---|
| Triage new errors | Lists recent issues and summarizes what is failing | Sentry issues |
| Check a release | Looks at errors tied to a specific release | Sentry releases |
| Spot slow spots | Reviews performance data for a project | Sentry performance |
| Explain an issue | Describes the error and where it appears | Sentry issue details |
This guide stops at the connection. For ready-made requests and scheduled routines, see Sentry with an AI Agent: Example Prompts and Routines.
What do I need before connecting?#
You need three things: a DoubleX Agentes account, a Sentry account with access to the projects you want to analyze, and a decision about who should be able to use the connection.
- A DoubleX Agentes account. Creating one is free and requires no credit card. During signup you share your company's segment, what you want to improve and where your information lives.
- Access in Sentry. The agent can only see what the Sentry account you authorize can see. Use an account that has access to the right projects, not necessarily the broadest one.
- An agent to use it. You can start from the diagnostic (a 5 to 10 minute conversation about your routine, free the first time), which turns into a team of agents with one click. Or create an agent on the Agents screen by describing what you need in your own words.
How do I connect Sentry on the Connectors screen?#
Open the Connectors screen, choose Sentry, and authorize it with your Sentry account. The whole flow takes a few steps in DoubleX Agentes:
- Sign in to DoubleX Agentes and open Connectors in the menu.
- Find Sentry in the list (it is in the development category) and start the connection.
- Sign in to Sentry when prompted and authorize access.
- Choose the scope of the connection: personal, team or company (see the next section).
- Open the agent that should use Sentry on the Agents screen and turn on the Sentry connector for it.
- Review which Sentry actions that agent is allowed to run (see below).
Credentials for connections are stored encrypted with AES-256-GCM, and each company has its own database, separate from other companies. Developers who prefer automation can also manage connectors through the API; see the Connectors API documentation.
Personal, team or company connection: which should I choose?#
Choose personal if only you should use your Sentry access, team if a group shares it, and company if every agent in the organization should reach Sentry through one shared connection.
| Scope | Who can use it | Good for |
|---|---|---|
| Personal | Only you | Your own investigation, an account tied to your permissions |
| Team | Members of a team | An engineering or support squad sharing the same projects |
| Company | The whole company | A shared monitoring setup that many agents use |
A practical rule: start narrow and widen later. A personal or team connection is easier to review than a company-wide one. Remember that routines run with the permissions of the person who created them, so create scheduled routines from an account whose Sentry access is appropriate.
How do I limit what the agent can do in Sentry?#
For each agent, you choose which systems it uses and, inside each system, which actions it may run. Actions that change data can stay turned off, leaving the agent read-only.
For a first setup, a sensible configuration is:
- Allow actions that read issues, releases and performance data.
- Leave off any action that changes something in Sentry until you have watched the agent work.
- Keep different permissions for different agents. A support agent might only need to read issues, while an engineering agent can have more.
Before the agent sends, deletes or changes anything in your systems, it asks for confirmation and shows the step-by-step of what it did. Every request also becomes a task with a full history on the Tasks screen: who asked, which tools the agent used and the result.
How do I test the connection with a first read-only request?#
Send a simple request that only reads data, then check the task history to confirm the agent actually used Sentry. A read-only test proves the connection works without changing anything.
Copy one of these to start:
- "List the five most recent unresolved issues in Sentry for our main project."
- "Which errors appeared in our latest release? Summarize the top ones."
- "Show me the Sentry issue with the most events this week and explain what is failing."
- "Give me a short overview of performance data for the checkout project."
After the answer, open the task on the Tasks screen. You should see the Sentry tool calls the agent made and the result it returned. If nothing from Sentry appears, check that the connector is turned on for that agent and that the Sentry account has access to the project you named.
Once the test passes, you can move to routines. If you want the agent to react on its own, the Triggers documentation explains schedules, webhooks and other ways to start a run. For another connector with the same setup flow, see How to Connect Jam to an AI Agent, which covers bug reports and console logs.
Frequently asked questions#
Do I need to write code to connect Sentry?#
No. You connect Sentry from the Connectors screen by authorizing your account, then turn it on for an agent. Code is only needed if you want to go further with the REST API, CLI or MCP server.
Can the agent change or delete things in Sentry?#
Only if you allow it. You choose which actions each agent can run, and actions that change data can stay off for read-only use. Before anything is sent, deleted or changed, the agent asks for confirmation.
Is my Sentry data used to train AI models?#
No. Your company's content is not used to train models, and requests only go to AI providers that do not use what they receive for training. You can also choose zero retention in the Models screen.
Can several agents use the same Sentry connection?#
Yes. A team or company connection can be turned on for more than one agent, each with its own permissions. A personal connection only works for the person who created it.
How do I know what the agent did in Sentry?#
Every request becomes a task with full history: the request, each tool the agent used and the result. You can review who asked, what was done and when.
Does it cost anything to try?#
Creating an account and running the first diagnostic is free, with no credit card. AI usage is paid with credits, and when your company's limit runs out, usage stops, so there are no surprise charges.
Related guides
- How to Connect Box to an AI Agent: Step-by-Step GuideLearn how to connect Box to an AI agent: set up the connector, limit it to read-only, choose who can use it, and check the history of your first request.
- How to Connect Jam to an AI Agent: Step-by-Step GuideLearn how to connect Jam bug reports to an AI agent in DoubleX Agentes: set up the connector, pick who can use it, limit permissions, and run a first test.