Sentry for AI agents

Sentry is an error tracking and application monitoring platform that surfaces crashes, performance issues, and releases across projects. Engineering teams use it to find and triage production problems. Connect it once through Arc0, and your agent, or Claude, ChatGPT and Cursor, can use it through one MCP endpoint, limited to what each user approved.

Developer toolsOAuth 2.0MCP + RESTsentry.io
AUDIT LOG · SENTRYPOLICY: acme-support
09:41:07 · claude · u_8f2read
sentry.list_project_tags
List project tags✓ allowed · 212ms
09:41:08 · claude · u_8f2read
sentry.get_alerts
Get alerts and monitors✓ allowed · 164ms
09:41:09 · claude · u_8f2write
sentry.create_organization_team
Create organization team✓ approved · approved by user
09:41:12 · claude · u_8f2destructive
sentry.delete_an_alert
Delete an alert✕ blocked · policy: deny
EVERY SENTRY CALL, ON THE RECORD
01 · USE CASES

What agents do in Sentry.

01

Review open issues before acting

Pull an organization's issues and their status before deciding which ones to bulk resolve or ignore.

02

Bulk resolve tracked issues

Mutate a batch of an organization's issues to mark them resolved once a fix ships, with approval first.

03

Record a new release

Create a release entry for an organization so incoming errors get tied to the right deploy.

02 · ACTIONS

211 Sentry actions, graded by risk.

Every Sentry action is tagged read, write or destructive, so one policy covers the whole app and new actions inherit the right default.

read

112

Look things up. Allowed by default.

  • sentry.get_alerts
    Get alerts and monitors
  • sentry.get_oldest_event
    Get oldest event
  • sentry.get_project_hook
    Get project hook
  • sentry.get_project_list
    Get project list
  • sentry.list_project_tags
    List project tags
  • sentry.get_project_events
    Get project events
  • sentry.list_organizations
    List organizations
  • sentry.list_project_users
    List project users
  • sentry.list_seer_ai_models
    List Seer AI models
  • sentry.get_project_event_stats
    Get project event stats
  • sentry.get_organization_details
    Get organization details
  • sentry.get_project_rule_details
    Get project rule details
  • sentry.get_organization_sessions
    Get organization sessions
  • sentry.get_project_monitor_by_id
    Get project monitor by id
  • sentry.get_scim_group_by_team_id
    Get SCIM group by team ID
  • sentry.list_organization_members
    List organization members
+ 96 MORE

write

64

Create and change things. Allow, or ask the user first.

  • sentry.update_a_monitor
    Update a monitor
  • sentry.update_user_email
    Update user email
  • sentry.add_team_to_project
    Add team to project
  • sentry.update_an_alert_by_id
    Update an alert by ID
  • sentry.update_a_monitor_by_id
    Update a monitor by ID
  • sentry.update_project_details
    Update project details
  • sentry.create_organization_team
    Create organization team
  • sentry.update_project_rule_by_id
    Update Project Rule By Id
  • sentry.update_monitor_for_project
    Update monitor for project
  • sentry.create_organization_monitor
    Create organization monitor
  • sentry.create_dashboard_with_widgets
    Create dashboard with widgets
  • sentry.create_release_deploy_for_org
    Create release deploy for org
  • sentry.update_environment_visibility
    Update environment visibility
  • sentry.update_organization_dashboard
    Update organization dashboard
  • sentry.create_organization_alert_rule
    Create organization alert rule
  • sentry.create_project_rule_for_alerts
    Create project rule for alerts
+ 48 MORE

destructive

35

Delete, cancel or archive. Ask first, or deny outright.

  • sentry.delete_an_alert
    Delete an alert
  • sentry.delete_bulk_alerts
    Delete bulk alerts
  • sentry.delete_project_key
    Delete project key
  • sentry.delete_project_hook
    Delete project hook
  • sentry.delete_project_rule
    Delete project rule
  • sentry.delete_project_by_id
    Delete project by id
  • sentry.delete_user_from_org
    Delete user from org
  • sentry.delete_project_issues
    Delete project issues
  • sentry.delete_project_replay
    Delete project replay
  • sentry.delete_project_monitor
    Delete project monitor
  • sentry.delete_member_from_team
    Delete member from team
  • sentry.delete_dsyms_for_project
    Delete dsyms for project
  • sentry.delete_user_emails_by_id
    Delete user emails by id
  • sentry.delete_organization_issue
    Delete organization issue
  • sentry.delete_release_file_by_id
    Delete release file by id
  • sentry.delete_external_team_by_id
    Delete an external team by id
+ 19 MORE
03 · HOW IT WORKS

Sentry in three steps.

  1. 01Your users connect SentryThey sign in to Sentry on Arc0 Connect, under your brand, and approve the access you ask for.
  2. 02You set the rulesReads run, writes like “create organization team” can wait for the user, and “delete an alert” can be denied outright.
  3. 03Any agent can actYour agent calls Sentry through the Arc0 SDK or MCP, and so can Claude, ChatGPT and Cursor. Every call lands on the audit log.
POLICY.TS
await arc0.policies.set('sentry', {
  read: 'allow',
  write: 'ask',        // create_organization_team
  destructive: 'deny',  // delete_an_alert
})

# Claude Code: the same connection, one URL
$ claude mcp add --transport http arc0 \
    https://mcp.arc0.ai/u/u_8f2
04 · AUTH AND DATA

How Sentry connects.

Users sign in to Sentry on Arc0 Connect and approve the scopes you request. Build with Arc0’s Sentry OAuth app, or bring your own so the Sentry consent screen names you. Tokens refresh automatically, and you can export them whenever you want.

The same Sentry connection serves your agent over MCP and your own backend over REST and the proxy, so a user connects once. How Arc0 handles credentials →

AUTH
OAuth 2.0
CREDENTIALS
Per-tenant encrypted vault
MODEL SEES
Results only, never credentials
AUDIT LOG
Every call, on every plan
05 · WORKS WITH

Use Sentry from any agent.

Claude
ChatGPT
Cursor
Codex
VS Code
OpenAI Agents SDK
Claude Agent SDK
Vercel AI SDK
Mastra
LangGraph
07 · FAQ

Sentry and Arc0, answered.

Q01

Can I use Sentry with Claude, ChatGPT or Cursor?

Yes. Connect Sentry to Arc0 once, then add your Arc0 MCP URL to Claude, ChatGPT, Cursor, Claude Code or any other remote-MCP client. Each assistant only gets the Sentry actions you allow.

Q02

How do users connect Sentry?

Users sign in to Sentry on Arc0 Connect and approve the scopes you request. Build with Arc0’s Sentry OAuth app, or bring your own so the Sentry consent screen names you. Tokens refresh automatically, and you can export them whenever you want.

Q03

Which Sentry actions can my agent take?

211 in total: 112 read, 64 write and 35 destructive, such as “create organization team”. Your policies decide which of them each agent may call.

Q04

Can I stop my agent from deleting things in Sentry?

Yes. Actions like “delete an alert” are graded destructive. Set destructive actions to deny, or to ask so the user approves each one, and blocked calls still show up on the audit log.

Q05

Can my own backend call Sentry too?

Yes. The same Sentry connection is available over REST and through the Arc0 proxy, so your product and your agent share one connection per user.

Get started

Plug Sentry into your agent.

Your users connect Sentry once, under your brand. Your agent gets 211 actions behind your policies, with every call on the record.

Free to build · MCP + REST · Audit log on every plan