Read Airbrake projects, error groups, notices, deploys and stats; mute or unmute error groups. For developers and on-call engineers who track application errors and deploys in Airbrake. Hosted by usefulapi — connect from Claude, ChatGPT, Cursor, or any MCP client.
Team and Enterprise: an Owner adds the connector for the organization first.
Run this command, then run /mcp in Claude Code to log in:
Add to Cursor or add this to ~/.cursor/mcp.json:
{
"mcpServers": {
"airbrake": {
"url": "https://airbrake.usefulapi.io/mcp"
}
}
}Add to VS Code or add this to .vscode/mcp.json:
{
"servers": {
"airbrake": {
"type": "http",
"url": "https://airbrake.usefulapi.io/mcp"
}
}
}No Add custom MCP server option? It depends on your account or workspace settings: see OpenAI's guide.
Run these commands. The second one opens the login page:
Other MCP clients (Windsurf, Cline, Zed and more): add the URL as a remote MCP server (Streamable HTTP). The client then opens the login page in your browser. Do not add an Authorization header: see Before you connect.
The AI lists a project's error groups, ordered by notice count or last notice, with error type, message, environment and counts, so you see what hurts most.
airbrake_list_projects airbrake_list_groups
The AI opens an error group with its full backtrace, then lists its notices with app version, user, URL and request params to show how it happens.
airbrake_get_group airbrake_list_notices
The AI reads the occurrence counts over time of one error group, or of a whole project, including notices dropped by rate limits or quota.
airbrake_get_group_stats airbrake_get_project_stats
The AI lists tracked deploys with environment, deployer, revision and version, and error groups filtered by deploy, so you can link errors to a release.
airbrake_list_deploys airbrake_list_groups airbrake_get_deploy
The AI mutes an error group, which stops its notifications while notices are still recorded, or unmutes it so notifications resume.
Changes data · airbrake_mute_group airbrake_unmute_group
The login page asks for your Airbrake User API key.
Find it in Airbrake → your avatar (top right) → Profile → User API key. Use the permanent User API key: a project key can only report errors, and a temporary user token expires. Your key is used only to call the Airbrake API on your behalf and to derive a private account id; it is never shown to anyone.
Add only the URL. Do not add an Authorization header or an API key to the client config. The server signs you in with OAuth: the login page asks for your Airbrake User API key. If the config has such a header, remove it: some clients then send that header instead of the login token, and every call fails with 401.
Turn on the Airbrake connector, then ask in plain words. The AI picks the tools.
| Tool | Type | What it does |
|---|---|---|
airbrake_list_projects | read | List projects List the Airbrake projects your user can see, with id, name, last deploy, notice total, rejected count, deploy count and resolved/unresolved error-group counts. Start here to get a project_id. Airbrake API: GET /api/v4/projects. |
airbrake_get_project | read | Get a project Get one Airbrake project: name, last deploy id/time, notice total, rejection count, deploy count and resolved/unresolved error-group counts. Airbrake API: GET /api/v4/projects/{project_id}. |
airbrake_list_groups | read | List error groups List a project's error groups ("errors" in the Airbrake app): each with id, resolved flag, error type/message/backtrace, context (environment, severity, component/action), last notice time, notice counts and last deploy. Sorted by last notice by default; order by notice_count, weight or created instead. Filter by deploy, archived or muted groups, or creation time. Backtraces are cut to the first 5 frames (set full_backtraces for all; airbrake_get_group shows one group in full). Airbrake API: GET /api/v4/projects/{project_id}/groups. |
airbrake_get_group | read | Get an error group Get one error group in full: resolved flag, every error with its complete backtrace (file, function, line, column), context, last deploy, last notice id/time, notice counts and creation time. Airbrake API: GET /api/v4/projects/{project_id}/groups/{group_id}. |
airbrake_get_group_stats | read | Get error group statistics Occurrence counts over time for one error group: parallel arrays `time`, `accepted` (notices counted) and `limited` (dropped by rate limits), aggregated per period, from `since` until `until` (default now). Use it to see whether an error is spiking or fading. Airbrake API: GET /api/v5/projects/{project_id}/groups/{group_id}/stats. |
airbrake_list_notices | read | List notices of an error group List the individual occurrences (notices) of one error group, newest first: each with id, deploy, errors and backtrace, context (environment, hostname, app version, user, URL, severity, notifier), environment, session and request params, and created time. Optionally only notices of one app version. Backtraces are cut to the first 10 frames unless full_backtraces is set. Airbrake API: GET /api/v4/projects/{project_id}/groups/{group_id}/notices. |
airbrake_get_notice_status | read | Get a notice's processing status Check what happened to a reported error, by the notice id that Airbrake's create-notice API returned: code = processed (groupId = its error group), rejected (message = why, e.g. an app-version filter), archived (past retention) or not_found (unknown or still processing). Airbrake API: GET /api/v4/projects/{project_id}/notice-status/{notice_id}. |
airbrake_list_deploys | read | List deploys List a project's tracked deploys: environment, username/email of the deployer, repository, revision (commit) and version. Use it to correlate new errors with a release. Airbrake API: GET /api/v4/projects/{project_id}/deploys. |
airbrake_get_deploy | read | Get a deploy Get one deploy: environment, deployer, repository, revision and version. Airbrake API: GET /api/v4/projects/{project_id}/deploys/{deploy_id}. |
airbrake_list_activities | read | List project activity List a project's activity feed: who did what and when (for example group.resolved), with user name, activity, the affected object (trackableType/trackableId) and timestamps. Airbrake API: GET /api/v4/projects/{project_id}/activities. |
airbrake_get_project_stats | read | Get project statistics Error volume over time for a whole project: parallel arrays `time`, `accepted` (notices counted), `limited` (dropped by rate limits) and `overQuota` (dropped over the plan quota), aggregated per period (1m, 5m, 15m, 30m, 2h, 3h, 6h, 12h, 24h). Airbrake API: GET /api/v5/projects/{project_id}/stats. |
airbrake_mute_group | write | Mute an error group WRITE: mute one error group. Airbrake removes it from the default error list and stops ALL notifications for it (email, Slack and other integrations) until it is unmuted. Notices are still recorded. Reversible with airbrake_unmute_group. Airbrake API: PUT /api/v4/projects/{project_id}/groups/{group_id}/muted. |
airbrake_unmute_group | write | Unmute an error group WRITE: unmute one error group, so it is back in the default error list and its notifications resume. Airbrake API: PUT /api/v4/projects/{project_id}/groups/{group_id}/unmuted. |
airbrake_usage_status | Usage status (free-tier meter) Report the caller's current usage: the plan, calls used this month, the limit of the plan (the free monthly cap or the Pro fair use limit), remaining, and whether the limit is reached. Read-only; does not count against the meter. | |
airbrake_request_feature | Request a missing feature Tell usefulapi that the user needs something the Airbrake tools cannot do yet (a missing tool, field or filter). Call it only when the user asks for something none of the tools can do, and tell the user that you send the request. Do not call it for errors or for normal requests. Do not include personal data, credentials or customer records. Stored with the product name and the client type (for example Claude or Cursor), without the user's account. At most 5 requests per day. Does not count against the meter. | |
airbrake_upgrade | Upgrade to Pro (unlimited*) Get Pro: unlimited tool calls (fair use limit: 10,000 a month) on Airbrake (the free tier caps monthly usage). Plans: "monthly" — $9/month; "yearly" — $90/year (2 months free). Cancel any time: Pro then runs to the end of the paid period, so it also suits a one-time job. *A fair use limit applies. For heavy use (near Pro's limit): "plus" or "plus_yearly" — Pro+, $39/month or $390/year, unlimited tool calls (fair use limit: 100,000 a month); an existing Pro subscription changes to Pro+ in place (needs confirm: true; the price difference for the rest of the period is charged at once). Otherwise returns a short link to open in your browser (it opens Stripe Checkout); after payment the account upgrades automatically. Does not count against the meter. | |
airbrake_cancel_subscription | Cancel the Pro subscription Cancel the caller's Airbrake Pro subscription at the end of the paid period (no refund for the current period; Pro continues until then, then the free tier applies). Requires confirm: true. Run airbrake_upgrade later to undo the cancel before the period ends. Does not count against the meter. |
| Plan | Price | Limit |
|---|---|---|
| Free | $0 | 50 tool calls / month |
| Proper user | $9/mo · $90/yr | Unlimited* |
*Pro fair use limit: 10,000 tool calls a month on this server. For more, write to [email protected].
Pro covers this Airbrake server only. Subscribe with airbrake_upgrade (it returns a Stripe Checkout link). Cancel any time with airbrake_cancel_subscription: Pro continues to the end of the paid period, with no refund for the current period, and running airbrake_upgrade before then undoes the cancel. Or write to [email protected].
Find the most frequent errors in a project. Inspect one error and its latest occurrences. Check if an error is spiking or fading. Match new errors to a recent deploy. Mute a noisy error or unmute it again. See what you can do for the details.
No. usefulapi is an independent service. It is not affiliated with or endorsed by Airbrake. The server calls the Airbrake API with your own Airbrake access, so it sees only the data that your account can see.
Yes. In Claude, add this server's URL as a custom connector. In ChatGPT, add it as a custom MCP server in Plugins. Then log in once. Cursor, VS Code, Claude Code and Codex work too. See the setup steps.
The login page asks for your Airbrake User API key. See Before you connect for where to find it.
No. Add only the URL. Do not add an Authorization header or an API key to the client config. The server signs you in with OAuth: the login page asks for your Airbrake User API key. If the config has such a header, remove it: some clients then send that header instead of the login token, and every call fails with 401.
The login stores them encrypted in the authorization grant of your connection. The server uses them to call the Airbrake API for you and to derive a private account id for usage metering. usefulapi does not show them on any page or in any reply. To stop all access, remove the connector and change or delete these credentials in Airbrake.
Yes, if you approve it. 2 of the 13 Airbrake tools can create or change data. The other 11 are marked read-only. Most MCP clients ask you to approve a tool call before it runs.
The Free plan gives 50 tool calls / month. Pro costs $9/mo or $90/yr, with unlimited tool calls (fair use limit: 10,000 tool calls a month), for this Airbrake server only. Cancel any time: Pro then runs to the end of the paid period. Run airbrake_usage_status to see how many calls you used.
Ask the AI to run airbrake_upgrade: it returns a Stripe Checkout link. To cancel, run airbrake_cancel_subscription. Pro continues to the end of the paid period.
Tell the AI what you wanted to do. It can send the request with airbrake_request_feature. We store the text with the server name, the type of AI client and the tool you tried, not with your account, and read every request when we plan new tools. You can send up to 5 requests per day.
Any client that supports remote MCP servers (Streamable HTTP) with OAuth login: Claude (web and desktop), Claude Code, ChatGPT, Codex, Cursor, VS Code, Windsurf and others.
This is a Model Context Protocol endpoint — meant to be connected from an AI client, not opened in a browser. An invalid_token response at the URL is the auth gate working as designed; clients authenticate automatically.