start using Sentry with your agents
sign up, open a project, and authorise Sentry from its connectors settings.
an alert tells you something broke, and that is the moment an agent should start looking. connecting sentry to an aldena project turns new issues, new errors and issue changes into runs: an agent creates an event routine on a sentry project or a single issue from a chat, and the next matching event starts a run with the error in hand, in that chat or in a new chat in the same project when the routine opens one per event. because each conversation in the aldena project gets its own sandbox, the agent can then go after the code the trace names. searching and changing issues goes through sentry's own mcp server.
subscribing from a chat
sentry_subscribe takes a resource, the events to watch, and what the agent should do when one arrives. a routine on a sentry project takes its slug such as acme-web and fires on new issues, resolved, assigned, ignored and unresolved issues, new errors and issue comments, including those on every issue inside that sentry project. a routine on an issue takes a short id such as ACME-WEB-42 and fires on the same events for that issue alone. sentry_unsubscribe takes an event routine off again, and the aldena project's Settings → Routines lists every routine the agents hold.
each notification carries the event routine, the actor, a one-line summary and a link. a new issue or a new error also carries the title, the culprit and the error message as sentry sends them, and a comment carries its text. that text is fenced off as untrusted data and capped at 4,000 characters, so an error message can inform the agent but cannot instruct it.
one organization, one aldena project
a sentry organization can be connected in one aldena project at a time, so an agent never hears about a neighbouring account's errors. sentry's consent screen shows the permissions aldena's integration is configured with, and disconnecting uninstalls the integration from the organization and frees it for another aldena project. when a connection needs a reconnect, aldena repairs it from the existing installation where it can.
going beyond triggers with sentry's mcp server
for searching issues and events, reading stack traces, and resolving or assigning issues, add sentry's mcp server to the same aldena project. it acts as the person who signs in to it.
- open the project's Settings → Connectors, and under MCP servers press Marketplace.
- install Sentry. it installs as
sentry-mcp, so it does not clash with the connector. - open the sign-in link it gives you and approve access in the browser.
once it is connected, agents reach its tools through mcp_search and mcp_call, each tool gets its own row in the project's permissions, and every action runs as the person who signed in, not as aldena. the mcp servers page covers the rest.
example tasks
- 01
watch the api project and look into every new issue.
the event routine on the sentry project starts a run in this chat for each new issue, with the title, culprit and message in the notification. the agent reads the code it points at in the conversation's sandbox and reports the likely cause.
sentry_subscribegithub_git_syncbash - 02
fix the checkout error and resolve it once the branch is pushed.
the agent reads the issue's events through sentry's mcp server, reproduces the failure in the conversation's sandbox, pushes a fix, and resolves the issue through the same server.
mcp_callgithub_git_syncbashgithub_git_push - 03
stop watching ACME-WEB-42.
the agent lists its routines, finds the event routine on that issue, and removes it, so changes there no longer start runs.
list_routinessentry_unsubscribe
what it will not do
the connector does not search issues or events, read stack traces, replays or profiles, or resolve, assign or delete anything. it reaches exactly one sentry organization, the one the aldena project connected, and no event starts a run while the connection waits for a reconnect.
what starts a run
every Sentry event an agent can subscribe to, grouped by resource. an agent subscribes from a chat with sentry_subscribe, and each matching event starts a run in that chat, or in a new chat in the same project when the routine opens one per event.
issue.createdissue.resolvedissue.assignedissue.ignoredissue.unresolvederror.createdcomment.created
issue.resolvedissue.assignedissue.ignoredissue.unresolvederror.createdcomment.created
what you grant
aldena's Sentry app is configured for this access, and Sentry shows it on its own consent screen. every project gets the same set, and there is nothing to pick.
- Projects (read)
Check the projects agents subscribe to.
- Issues and events (read)
Check the issues agents subscribe to.
how to connect Sentry
connecting Sentry takes one authorisation round trip, started from inside the project that needs it.
- open the project's connectors settings
connections belong to a project, not to your account or the team. open the project that needs Sentry, go to settings → connectors, and press connect. other projects stay untouched.
- authorise with a real account
any member of the project can start the connection, and they authorise as themselves. Sentry asks only for the fixed access listed above, with no scopes to choose.
- an agent subscribes from a chat
ask an agent in a project chat to watch something in Sentry. it calls sentry_subscribe with the resource, events and intent, and settings → routines lists each routine it holds.
- a matching event starts a run
each matching event starts a run with the event and its content in hand, in the chat where the agent subscribed or in a new chat in the same project when the routine opens one per event. disconnecting Sentry stops every subscription to it.
data handling and security
a connector holds one credential and turns Sentry events into runs. the credential stays inside the project that connected it, and event text is data, never instructions.
- one connection, one project
the Sentry credential is stored against the project it was authorised in. no agent in another project can use it, and there is no team-wide connection to inherit by accident.
- tokens encrypted at rest
the Sentry token is encrypted before it is stored, and it never appears in a chat or in an agent's prompt. agents get the events, not the credential.
- events only where a routine watches
aldena verifies each delivery came from Sentry. an event starts a run only for a routine it matches, in the chat that set the routine up or in a new chat in the same project when the routine opens one per event. nothing starts while the connection needs a reconnect.
- event content is untrusted
event text may come from a stranger. it reaches the agent as fenced data capped at 4,000 characters, and the agent is told never to follow instructions in it.
- revocation is one click
disconnect Sentry and the project's event routines on it stop at once, along with the stored credential.

each Sentry tool agents hold has its own allow, ask or deny row in the project's permissions. see human in the loop for approvals and isolated projects for isolation.
faq
can an ai agent triage sentry issues?
yes, in two parts. the sentry connector starts the run: a new issue or error in a watched sentry project wakes the agent with the title, culprit and error message. searching, reading events and resolving or assigning issues goes through sentry's own mcp server, added from the aldena project's mcp marketplace with a browser sign-in.
does it stop at reading, or can it fix the bug?
each conversation in the aldena project gets its own sandbox, so the agent a sentry event woke up can clone the repository onto it, reproduce the failure over bash, write the change, run the tests and push a branch. that is the step a monitoring assistant cannot take.
what does a sentry event notification contain?
the event routine, the actor, a one-line summary and a link. a new issue or a new error also carries the title, the culprit and the error message as sentry sends them, and a comment carries its text. that text is fenced off as untrusted data and capped at 4,000 characters.
should an agent subscribe to every error?
usually not. error.created fires for every single error event, not once per issue, so on a busy sentry project it is very high volume. issue.created fires once when a new kind of failure appears, which is what most projects want to wake on.
can two aldena projects connect the same sentry organization?
no. a sentry organization can be connected in one aldena project at a time, so an aldena project only ever hears about the sentry projects of the organization it connected. disconnecting uninstalls the aldena integration from that organization and frees it for another aldena project.
spin up your first project.
one project per client, product, or piece of work, staffed with intake, requirements, build and code review agents.
