developer-toolsapi_key

Bugsnag

Bugsnag is an error monitoring and stability management platform that helps developers identify, prioritize, and fix software bugs.

Verdict

Bugsnag tracks errors and stability across your production apps. @mention it in Switchy to query crash reports, triage issues by severity or user impact, bulk-update error statuses, and configure alerting integrations — all without leaving the conversation. Engineers and support leads get the most value: they can pull context on a specific bug, mark batches resolved after a deploy, or set up custom event fields to filter by user cohort. Auth requires an API key with project-level access; read-only keys won't let you update errors or manage integrations.

Common use cases

Triage crashes by user impact before standup
Bulk-resolve errors after a hotfix deploy
Create saved searches for high-severity regressions
Configure Slack alerts for new error spikes
Query custom event fields by user cohort

Integration

Vendor: Bugsnag
Category: developer-tools
Auth: API_KEY
Tools: 33
Composio slug: bugsnag

Tools

Bulk Update Errors
Tool to bulk update multiple errors in a project. use when batching identical updates across many errors.
Configure Integration
Tool to configure a new integration for a bugsnag project. use after selecting integration key and preparing credentials.
Create Custom Event Field
Tool to create a custom event field for a project. use after confirming the project id to add a new filterable field. example: "create a field 'user.id' mapping to 'metadata.user.id' for project 515fb..."
Create Saved Search
Tool to create a new saved search for a project. use when you need to persist a set of filters and sort order as a saved search for a project after confirming project details.
Delete a configured integration
destructive
Tool to delete a configured integration. use when you need to remove an existing integration from a project after confirming the integration id. this action cannot be undone and requires organization administrator or project owner permissio
Delete All Errors in a Project
destructive
Tool to delete all errors in a project. use when you need to remove all error and event data for a given project. use with caution after confirming the project id, as this action is irreversible and requires project owner or organization ad
Delete Collaborator
destructive
Tool to remove a collaborator from an organization. use when revoking collaborator access after confirming both organization and collaborator ids. requires organization administrator privileges.
Delete Event Field
destructive
Tool to delete a custom event field. use when you need to remove a specific event field after confirming the project id and display id.
Delete Project
destructive
Tool to delete a project. use when you need to permanently remove a project after confirming the project id. requires appropriate permissions.
Delete Saved Search
destructive
Tool to delete a saved search. use when you need to permanently remove a saved filterset by its id after confirming the saved search identifier.
Get Saved Search
Tool to retrieve a saved search by id. use when you need to fetch details of a saved search after confirming its id.
Get Saved Search Usage Summary
Tool to get usage summary for a saved search. use after confirming the saved search id. example: "get usage summary for saved search 515fb9337c1074f6fd000003".
Invite Collaborator
Tool to invite a collaborator to an organization. use when you need to add a new user by email to grant them access. ensure the organization id is correct before calling.
List Collaborator Access Details for Projects
Tool to list summary details of the projects a collaborator has access to. use when reviewing a collaborator's access across an organization.
List Collaborators
Tool to list collaborators in an organization. use when you need to retrieve all collaborators for a specified organization. example: "list collaborators for organization 515fb9337c1074f6fd000001".
List Collaborators on Project
Tool to list collaborators on a project. use when you need to retrieve all users with access to a given project after confirming its id. example: "list collaborators for project 515fb9337c1074f6fd000001".
List Configured Integrations for Project
Tool to list configured integrations for a project. use after confirming the project id to retrieve all integration summaries.
List Errors on Project
Tool to list all errors in a project. use when you need to retrieve errors with optional filtering by version and sorting.
List Event Fields for Project
Tool to list event fields for a project. use when you need to discover all available fields for filtering events in a specified project. example: "list event fields for project 515fb9337c1074f6fd000001".
List Events on Project
Tool to list events for a project. use when you need to retrieve all error occurrences (events) for a given project after confirming its id. example: "list events for project 515fb9337c1074f6fd000003".
List Organizations
Tool to list organizations for the authenticated user. use after validating the api token when you need to retrieve all organizations the current user has access to.
List Pivots for Project
Tool to list pivots for a project. use when you need to retrieve all pivot definitions available in a specified project.
List Projects
Tool to list projects in an organization. use when you need to retrieve all projects under a specified bugsnag organization after confirming the organization id.
List Release Groups
Tool to list release groups for a project. use when you need to retrieve all release groups of a specific bugsnag project after confirming the project id.
List Releases
Tool to list releases for a project. use when you need to retrieve all releases of a specific bugsnag project after confirming the project id.
List Saved Searches on Project
Tool to list saved searches for a project. use after confirming project id to retrieve all saved searches, optionally filtering by shared flag.
List Supported Integrations
Tool to list all supported integrations. use when discovering available integration services before configuring a project.
List Trends for Project Buckets
Tool to list trend buckets for a project. use when you need time-segmented event counts to analyze occurrence patterns.
Regenerate Project API Key
Tool to regenerate a project's api key. use after confirming the project id and when you need to rotate the notifier api key.
Show Collaborator Access Details for Project
Tool to show a collaborator's access details for a project. use after confirming organization, collaborator, and project ids have been obtained and you have access rights.
Show Collaborator on Organization
Tool to show a collaborator in an organization. use after confirming organization and collaborator ids; requires organization administrator access.
Show Collaborator on Project
Tool to show a collaborator in a project. use when you need detailed information about a specific collaborator after confirming both project and collaborator ids.
Update Collaborator Permissions
Tool to update a collaborator's project access permissions. use when you need to change which projects a collaborator can access or their roles within each project. supply only project ids or project roles.

Setup

Setup guide

11. In Switchy, open your workspace settings and click 'Add Integration' under the MCP section. 2. Select Bugsnag from the developer tools category. 3. In a separate tab, log into your Bugsnag account and navigate to Settings → My Account → Personal Auth Tokens. 4. Generate a new token with at least 'Read' scope for projects you want to query; add 'Write' scope if you plan to update errors or configure integrations. 5. Copy the token and paste it into Switchy's API key field, then click 'Connect'. 6. Switchy will verify the token and list your accessible organizations and projects. 7. To test, open any Space and type '@Bugsnag list recent errors in project [your-project-id]' — you should see a summary of open issues. 8. If the response is empty or shows an auth error, confirm your token has the correct scopes and hasn't expired.

What teammates see: by default, memories from Bugsnag are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.

Works well with

Top models

Compatibility data appears once enough Spaces have used this MCP together with a given model.

How Switchy teams use it

Not enough Spaces yet to publish anonymised usage stats (we require ≥ 50 Spaces per week).

Starter prompts

Recent Critical Errors

@Bugsnag show me the most recent critical errors in project abc123, sorted by event count descending

Open in a Space →

Bulk Resolve After Deploy

@Bugsnag bulk update errors [error-id-1, error-id-2, error-id-3] in project abc123 and set status to resolved

Open in a Space →

Create Custom Event Field

@Bugsnag create a custom event field 'user_tier' mapping to 'metadata.account.tier' for project abc123

Open in a Space →

Save High-Impact Search

@Bugsnag create a saved search named 'High User Impact' in project abc123 filtering for errors with user count greater than 100

Open in a Space →

Configure Slack Integration

@Bugsnag configure a Slack integration for project abc123 using webhook URL https://hooks.slack.com/services/YOUR/WEBHOOK/PATH

Open in a Space →

Example outputs

Illustrative - representative of the model's voice and quality, not literal recordings.

Prompt

@bugsnag show me all unresolved errors in production from the last 7 days, grouped by error class

Output

Here's a summary of unresolved production errors from the past week: **RuntimeError** (47 occurrences, 23 users affected) - Most recent: 2 hours ago - Top context: payment processing flow **NullPointerException** (31 occurrences, 12 users affected) - Most recent: 4 hours ago - Top context: user profile loading **NetworkTimeoutException** (18 occurrences, 8 users affected) - Most recent: 1 hour ago - Top context: third-party API calls This illustrative example shows how Bugsnag would return aggregated error data with occurrence counts and affected user metrics.

Notes

This read-heavy example demonstrates Bugsnag's core error monitoring capability. The MCP requires an API key with read access to the specified project. Response shape and grouping depend on your project's error volume and custom metadata configuration. Rate limits apply to bulk queries across large time windows.

Prompt

@bugsnag mark all 'NetworkTimeoutException' errors in the checkout project as resolved and add a comment that we deployed the retry logic fix

Output

Bulk update completed for checkout project: ✓ Updated 18 errors matching 'NetworkTimeoutException' ✓ Status changed to: Resolved ✓ Comment added: "Deployed retry logic fix per PR #847" ✓ Resolution timestamp: 2024-01-15 14:32 UTC All affected error instances now show as resolved in your Bugsnag dashboard. Users will no longer receive notifications for these specific errors unless they reoccur with different stack traces. This illustrative example shows a typical bulk update confirmation.

Notes

This write-action example uses the Bulk Update Errors tool to modify error states across multiple instances. Requires API key with write permissions. Side-effect: This permanently changes error status and triggers notifications to configured integrations (Slack, Jira, etc.). Cannot be undone — verify the error filter before executing.

Prompt

@bugsnag analyze the top 5 errors from this week and draft a status update email explaining which ones are fixed, which are in progress, and what the user impact has been

Output

Based on this week's error data, here's a draft status update: **Resolved (2 errors, 65 total occurrences)** NetworkTimeoutException and CacheInvalidationError are now fixed in production. These affected 31 users during checkout flows but have seen zero occurrences since yesterday's deployment. **In Progress (2 errors, 42 occurrences)** AuthTokenExpiry and ImageUploadFailure are under investigation. Combined impact: 19 users, primarily in profile editing. Fix targeted for Friday release. **Monitoring (1 error, 8 occurrences)** RateLimitExceeded appears intermittent with only 3 affected users. Tracking for patterns before prioritizing. This illustrative synthesis combines error metrics with AI-generated narrative.

Notes

This synthesis example pairs Bugsnag's error data with AI reasoning to create human-readable summaries. The AI interprets occurrence counts, affected users, and timestamps to categorize urgency. Accuracy depends on error metadata quality and recency of the data fetch. Always verify critical details before sending to stakeholders.

Use-case deep-dives

Post-incident error cleanup

When Bugsnag MCP handles mass error triage after a bad deploy

A 6-person engineering team ships a config bug that floods Bugsnag with 2,000 duplicate errors across three projects. The on-call engineer needs to bulk-close the noise, preserve the real issues, and document the incident pattern. The Bugsnag MCP wins here because it exposes bulk operations and saved searches that the web UI buries under pagination. Use 'Bulk Update Errors' to mark the duplicates resolved in one call, then 'Create Saved Search' to pin the filter for post-mortem review. The threshold: if your team triages fewer than 50 errors per incident, the web UI is faster. Above that, the MCP pays off by letting you script the cleanup and share the query with the team in Switchy.

Customer support error lookup

Why Bugsnag MCP falls short for support ticket workflows

A 3-person support team fields tickets where customers report vague errors ('the app crashed on checkout'). They want to search Bugsnag by user ID, pull the stack trace, and paste it into the ticket thread. The Bugsnag MCP doesn't win this scenario because it lacks a direct 'search errors by custom field' tool—you'd need to create a saved search first, then list errors, then filter client-side. That's three round-trips for a task the Bugsnag web UI handles in one search bar. The MCP's 33 tools skew toward admin tasks (integrations, collaborators, bulk updates), not ad-hoc error lookup. If your support flow is 'search once, copy stack trace', stay in the Bugsnag UI.

Weekly error dashboard prep

When Bugsnag MCP automates recurring error reporting for standups

A 5-person product team runs a Monday standup where the PM reviews last week's top errors and assigns owners. The ritual takes 15 minutes of clicking through Bugsnag projects, screenshotting counts, and pasting into Notion. The Bugsnag MCP turns this into a 2-minute Switchy prompt: 'List errors for project X from last 7 days, group by severity, create a saved search for P0s.' The MCP's strength is repeatable queries—once you script the report, the team reuses it every week without re-learning the UI. The trade-off: setup takes 10 minutes the first time to map project IDs and confirm field names. If your standup cadence is monthly or ad-hoc, the setup cost exceeds the payoff.

Frequently asked

What does the Bugsnag MCP do in Switchy?

It connects your Bugsnag error monitoring to Switchy's AI workspace so you can query crash reports, bulk-update errors, manage saved searches, and configure integrations without leaving the conversation. The MCP exposes 33 tools covering everything from reading error details to deleting entire projects' worth of data. You ask in natural language; the AI calls the right Bugsnag API endpoints.

Do I need admin access to connect Bugsnag?

You need a Bugsnag API key with permissions matching what you want to do. Read-only keys work for querying errors and viewing projects. Destructive operations—deleting collaborators, wiping all errors in a project, removing integrations—require organization administrator privileges. Generate the key in Bugsnag's settings and paste it into Switchy's connection flow.

Can the Bugsnag MCP delete errors or entire projects?

Yes. The MCP includes tools like Delete All Errors in a Project and Delete a Configured Integration. These are irreversible and require careful confirmation of project or integration IDs. The AI will prompt you before executing destructive commands, but the underlying capability is there—useful for cleaning up test data or revoking stale integrations.

How is this different from using Bugsnag's dashboard directly?

The dashboard is faster for one-off lookups and visual triage. The MCP shines when you want to bulk-update 50 errors with the same status, create custom event fields across multiple projects, or cross-reference Bugsnag data with logs from another tool—all in one Switchy thread. You trade Bugsnag's charts for conversational batch operations and multi-tool context.

Who on the team should connect Bugsnag to Switchy?

Whoever owns on-call rotation or triages production errors. If your team shares a Switchy workspace, one person's Bugsnag connection lets everyone query errors and create saved searches. For destructive actions—deleting collaborators, wiping projects—only connect an API key with admin scope, and limit that connection to senior engineers who understand the blast radius.