BugSnag MCP server
BugSnag in any MCP client, through Memtro: BugSnag does not publish this server. Memtro's BugSnag connector reads BugSnag through its API with the credentials you connect, and one endpoint hands those tools, every other system you connect and Memtro's memory tools to Claude Code, Cursor, Claude Desktop or anything else that speaks Streamable HTTP, as you.
What the model can do
BugSnag error monitoring: projects and their open error counts, errors by last seen, first seen, users or events, an error with its recent events and where in the code, comments on an error, releases and a project's people. Can comment on an error, fix, reopen, ignore, snooze, assign or re-prioritise errors, link or create an issue, discard or permanently delete an error, and edit or delete a comment.
Reads
bugsnag_list_projectsBugSnag projects in an organization (the connection's first one by default): id (needed by the other tools), name, type, language…bugsnag_list_errorsErrors in a BugSnag project, sorted by last seen (default), first seen, users affected or events, newest or most first unless asked otherwise: class…bugsnag_get_errorOne BugSnag error with its most recent events: when each happened, app version and stage, handled or not, and the first in-project stack frame.bugsnag_list_commentsComments on a BugSnag error, newest first: who wrote it, when, and what.bugsnag_list_releasesRecent releases of a BugSnag project, newest first: version, stage, when released, by whom, errors introduced and seen…bugsnag_list_collaboratorsPeople with access to a BugSnag project: id (to assign an error to) and name.
Acts, when you ask
These change something in BugSnag. The model is told to use them only when you ask for exactly that, and each one is granted by name: per team on a shared connection, by the owner on a personal one.
bugsnag_add_commentAdds a comment to a BugSnag error, written as the user whose token this is, for the team working on it to read.bugsnag_update_errorChanges a BugSnag error's workflow: fix, open (reopen), ignore (events are still collected but no notifications are sent)…bugsnag_update_errorsApplies one workflow change (fix, open, ignore, snooze, assign, severity, link or unlink an issue, undiscard…bugsnag_discard_errorDiscards future events for a BugSnag error: the error and its existing events stay, but new occurrences are not stored or counted…bugsnag_delete_errorPermanently deletes a BugSnag error and all its events.bugsnag_update_commentReplaces the text of a BugSnag comment.bugsnag_delete_commentPermanently deletes a BugSnag comment.
Connect it
First connect BugSnag on Dashboard → Connections, then add Memtro to your client once. The BugSnag tools appear alongside the tools of everything else you have connected; nothing is configured per connector.
Claude Code
The Memtro plugin registers the server and a skill for when to use it; the tools are then available in every session.
claude plugin marketplace add memtro-org/claude-memtro
claude plugin install memtro@memtro --config api_key=sk-memtro-...Cursor
Add the server to .cursor/mcp.json in the project, or to the global file under ~/.cursor/.
{
"mcpServers": {
"memtro": {
"url": "https://platform.memtro.com/api/mcp",
"headers": {
"Authorization": "Bearer sk-memtro-..."
}
}
}
}Claude Desktop
Create a personal API key on Dashboard → Profile → API keys. In Claude Desktop add an MCP server of type HTTP with URL https://platform.memtro.com/api/mcp and the header Authorization: Bearer sk-memtro-…
OpenCode
A remote server in opencode.json.
{
"mcp": {
"memtro": {
"type": "remote",
"url": "https://platform.memtro.com/api/mcp",
"headers": {
"Authorization": "Bearer sk-memtro-..."
}
}
}
}Pi
One command: pi mcp add memtro --url https://platform.memtro.com/api/mcp --bearer-token-env-var MEMTRO_API_KEY. Tools arrive as mcp__memtro__*. The pi-memtro extension registers it for you, alongside Memtro as the model.
Any other client
Add an MCP server of type HTTP (Streamable HTTP, stateless) at https://platform.memtro.com/api/mcp with the header Authorization: Bearer sk-memtro-.... Codex, Zed, Windsurf, Cline and the SDKs all take a remote server this way.
Who may use it
The key identifies the person, and the server offers that person the tools they may use everywhere else in Memtro. A shared BugSnag connection is used by the teams an administrator granted it to, tool by tool, on the Permissions page; a personal connection is used by its owner alone, and its tools that act stay off until the owner allows them. See Teams and permissions and Connectors.
Ask it like this
- › What new errors appeared since the last release?
- › Which errors affected the most users this week?
- › Show me the recent events for the checkout timeout error
- › Mark the login crash as fixed and assign the payment webhook error to Priya
- › Add a comment to the login crash saying the fix ships in 4.12
Questions
- Is this an official BugSnag MCP server?
- No. It is Memtro's MCP server with Memtro's BugSnag connector behind it: the connector reads BugSnag through its API with the credentials you connect on Dashboard → Connections, and BugSnag publishes nothing here. The same endpoint carries every other system you connect and Memtro's memory tools.
- What does it cost?
- Nothing on top of Memtro. One price per organisation by its number of users, from $50 a month for up to 10 users after a 7-day free trial, with every connector, the MCP server, jobs and memory included. You bring your own model keys.
- Can it change anything in BugSnag?
- Only when you ask, and only what it has been granted: add a comment to an error, written as the token's user; fix, reopen, ignore or snooze an error; assign an error to a person or team, or unassign it; change an error's severity; link an error to an external issue, unlink it, or create one in the configured tracker; apply one of those changes to up to 25 named errors; discard future events for an error; permanently delete an error and its events; edit a comment written as the token's user; permanently delete a comment. Every other tool reads, searches or lists. An administrator grants each acting tool by name per team, and a personal connection acts only after its owner allows it.
- Which clients work?
- Any MCP client that speaks Streamable HTTP: Claude Code, Cursor, Claude Desktop, OpenCode, Codex, Zed, Windsurf, Cline and the Anthropic and OpenAI SDKs. There is no local process to run; the server is remote and authenticated with your key.