Honeybadger MCP server
62 operations0 webhooks
Error tracking, uptime monitoring, check-ins, deploys and Insights queries from Honeybadger's v3 API.
Try Honeybadger with your agent
Connect Honeybadger, then paste this into Claude, ChatGPT or Cursor. For ideas from other apps, see the use cases.
Using Corsair, show me what you can do in Honeybadger on my account, then run one read-only example and show me the result.
Honeybadger tools and triggers
Tools are what your agent can do in Honeybadger through Corsair's MCP server. Triggers are the Honeybadger events a routine can start from.
Faults
- Affected UsersRead-only
Read the 500 users a fault affected most, ranked by notice count
faults.affectedUsers
- AssignWrite
Assign a fault to a project member, who is notified of it
faults.assign
- DeleteDestructive
Permanently delete a fault; its notices stay readable by ID until retention removes them
faults.delete
- GetRead-only
Read one fault by ID; a merged fault redirects to the fault it was merged into
faults.get
- IgnoreWrite
Ignore faults, which stops data collection for them, by ID or by query with `all: true`
faults.ignore
- ListRead-only
List faults with error search, sorting and time filters; occurrence counts cover the requested window
faults.list
- MergeDestructive
Merge one fault into another in the background; the merged fault keeps its tags, assignee and state on neither side, and its ID redirects for the retention period
faults.merge
- OccurrencesRead-only
Read occurrence counts for one fault over a window, one bucket per interval including empty ones
faults.occurrences
- Pause RecordingWrite
Stop recording a fault for an hour, a day or a week, discarding notices during the pause
faults.pauseRecording
- ResolveWrite
Resolve named faults, or every fault a query and time filter match with `all: true`; `dry_run` reports what would change
faults.resolve
- Resume RecordingWrite
Resume recording a fault; a fault that is not paused is unchanged
faults.resumeRecording
- SnoozeWrite
Snooze notifications for a duration or an occurrence count, replacing any existing snooze
faults.snooze
- SummaryRead-only
Read fault counts grouped by environment and resolved or ignored state, with the same filters as the fault list
faults.summary
- UnassignWrite
Remove the assignment from a fault, leaving it unowned
faults.unassign
- UnignoreWrite
Stop ignoring faults, so collection resumes, by ID or by query with `all: true`
faults.unignore
- UnresolveWrite
Mark faults unresolved, by ID or by query with `all: true`; `dry_run` reports what would change
faults.unresolve
- UnsnoozeWrite
Remove a fault's snooze, so notifications resume from the next occurrence
faults.unsnooze
- UpdateWrite
Change a fault resolved or ignored state, assignee, tags, or whether it resolves on the next deploy
faults.update
Check Ins
- CreateWrite
Create a check-in on a simple period or a cron schedule; a simple schedule needs `report_period` and a cron schedule needs `cron_schedule`
checkIns.create
- DeleteDestructive
Delete a check-in, which stops monitoring that scheduled job
checkIns.delete
- EventsRead-only
List a check-in's state changes newest first, paging backwards on a timestamp
checkIns.events
- GetRead-only
Read one check-in with its schedule, last report and next expected report
checkIns.get
- ListRead-only
List a project's check-ins with their state and next expected report
checkIns.list
- ReplaceDestructive
Replace the project's whole set of check-ins by slug; every check-in the list does not name is deleted
checkIns.replace
- UpdateWrite
Change a check-in's name, slug, schedule or grace period; switching schedule type clears the other one's fields
checkIns.update
Environments
- Bulk DeleteDestructive
Delete several environments at once; if any ID is not one of the project's, nothing is deleted
environments.bulkDelete
- Bulk UpdateWrite
Turn notifications on or off for several environments at once; if any ID is not one of the project's, nothing changes
environments.bulkUpdate
- CreateWrite
Create an environment; names are unique per project and case-sensitive
environments.create
- DeleteDestructive
Delete an environment; its faults and notices stay, and the next notice under that name recreates it with notifications on
environments.delete
- GetRead-only
Read one environment and whether its errors send notifications
environments.get
- ListRead-only
List a project's environments and whether each sends notifications
environments.list
- UpdateWrite
Change whether an environment notifies; an environment cannot be renamed
environments.update
Projects
- CreateWrite
Create a project, optionally in a team; `team_id` requires `team:write` and rejects the whole request without it
projects.create
- DeleteDestructive
Permanently delete a project and all its data
projects.delete
- GetRead-only
Read one project by ID, including its fault counts and environment names
projects.get
- ListRead-only
List projects the token can reach, optionally filtered by exact name, with paging
projects.list
- OccurrencesRead-only
Read notice counts for one project over an hour, day, week or month, one count per interval
projects.occurrences
- StatsRead-only
Read a project fault and notice counts for the last day, week and month
projects.stats
- UpdateWrite
Change a project's settings; a null clears a nullable field and a field left out keeps its value
projects.update
Sites
- ChecksRead-only
List a site's individual uptime check results, newest page first
sites.checks
- CreateWrite
Create an uptime site; `match` is required for every match type except `success`, which refuses it, and `timeout` must be shorter than the check interval
sites.create
- DeleteDestructive
Delete an uptime site and stop monitoring it
sites.delete
- GetRead-only
Read one uptime site with its URL, match rule and current state
sites.get
- ListRead-only
List a project's uptime monitoring sites with their URLs and current state
sites.list
- OutagesRead-only
List a site's outage periods, with the status and reason each failed
sites.outages
- UpdateWrite
Change an uptime site's URL, matching, headers, locations, frequency or active state
sites.update
Comments
- CreateWrite
Comment on a fault; `@mentions` notify the named project members
comments.create
- DeleteDestructive
Permanently delete a comment from a fault
comments.delete
- GetRead-only
Read one comment on a fault, with its author and body
comments.get
- ListRead-only
List a fault's comments newest first, paging backwards with a cursor
comments.list
- UpdateWrite
Replace the text of a comment on a fault; other comment fields cannot be changed
comments.update
Deploys
- DeleteDestructive
Delete a deploy record; the deploy is not re-read anywhere
deploys.delete
- GetRead-only
Read one deploy record, with its environment, revision and author
deploys.get
- ListRead-only
List deploys oldest-page-first with cursors, optionally by exact environment or deploy username
deploys.list
Account
- GetRead-only
Read the account the token belongs to, with its name and primary email
account.get
- UsageRead-only
Read the account quota and how much of it is used, for errors, uptime and check-ins
account.usage
Notices
- GetRead-only
Read one notice by its UUID, without naming its project or fault
notices.get
- ListRead-only
List a fault's notices newest first, with source-mapped backtraces unless `unmapped` is set
notices.list
Insights
- QueryRead-only
Run a BadgerQL query over the project event streams and return its rows; a POST that changes nothing
insights.query
Occurrences
- ListRead-only
Read notice counts for every project the token can see, ordered by project name
occurrences.list
Streams
- ListRead-only
List event streams; the ID is what `insights.query` takes in `stream_ids`
streams.list
Token
- GetRead-only
Report what the current token is: its kind, scopes, account and the projects it can reach. Needs no scope
token.get
Connect Honeybadger to Claude, ChatGPT or Cursor
- 1
Connect Honeybadger in Corsair
Add your Honeybadger API key once in Corsair. Corsair stores the credential encrypted and makes the Honeybadger API calls for your agent, so it never goes into a prompt or a config file. Connect Honeybadger
If your link lets the agent add apps, you can skip this: it gives you a sign-in link the first time it needs Honeybadger. On a ChatGPT personal plan, connect Honeybadger here first.
- 2
Add Corsair's MCP server to your agent
Every client uses the same address:
https://mcp.corsair.dev/mcpOpen Customize → Connectors → Add custom connector, paste the address, then select Connect.Add a custom MCP server with the address and choose OAuth. Personal plans get read tools only.Add to CursorOr add this to~/.cursor/mcp.json. Cursor opens Corsair to sign in the first time it connects.{ "mcpServers": { "corsair": { "url": "https://mcp.corsair.dev/mcp" } } }Run this, then run/mcpin Claude Code to sign in.claude mcp add --transport http corsair https://mcp.corsair.dev/mcpCodex, Gemini CLI, VS Code, Zed and other clients with OAuth sign-in work the same way. Clients without it, such as GitHub Copilot's coding agent, use a link's own address and an API key. Setup for every client
- 3
Sign in and choose what needs your approval
Your client opens Corsair to sign in. Pick a link: the apps your agent can use and a mode that decides what it may do without asking. In the default Cautious mode, Honeybadger's read and write operations run right away and its 10 destructive operations wait for your approval. Strict mode asks before every write and refuses destructive operations, and Read mode allows reads only.
FAQ
Install corsair and @corsair-dev/honeybadger, add the Honeybadger plugin, and connect with an API key. Then call methods like corsair.honeybadger.api.account.get() from your app. Corsair keeps each tenant isolated.
Yes. Honeybadger requires an API key from your Honeybadger account or developer dashboard. Create credentials with access that matches what you automate, then paste them into Corsair or connect through managed auth in Hub.
Yes. One Corsair client can hit Honeybadger, GitHub, Notion, Stripe, and the rest. Trigger Honeybadger from a webhook, push updates to Slack, or chain actions across plugins. Same tenant context across integrations, plain TypeScript.
Teams use Corsair for error tracking, uptime monitoring, check-ins, deploys and Insights queries from Honeybadger's v3 API. Common builds include sync jobs, admin tooling, and agents that call Honeybadger APIs from your backend.
The SDK is open source. Self-host and you pay Corsair nothing for the integration layer. Hub bills for hosted auth and execution, not per Honeybadger API call. Hub has a generous free tier. For hobby projects you can probably run the whole thing on free. Paid plans start to matter with more tenants or heavier traffic.
We maintain @corsair-dev/honeybadger in the open repo. When Honeybadger ships API changes, upgrade the package. Your call sites stay typed. You should not need to rewrite the integration layer each time.
No. Corsair does not read your Honeybadger data or hold individual access credentials. Credentials stay in your database. API calls run on your instance. We do not store Honeybadger data on your behalf.
Yes. Add https://mcp.corsair.dev/mcp as a custom connector or MCP server in Claude, ChatGPT, Cursor, VS Code, Claude Code, Codex or any client that supports remote MCP servers with OAuth sign-in, then sign in to Corsair. On ChatGPT personal plans, custom servers currently get read tools only.
Yes. Put Honeybadger on a link in Strict mode and every write waits for your approval, while destructive actions are refused. The default Cautious mode asks only before the 10 destructive Honeybadger operations. You can change a link's mode at hub.corsair.dev/me.