Bugsnag connector for AI agents

CALL ANY TOOL

List and inspect errors, pull events and stack traces, check release stability, and resolve or reopen errors, each call scoped to the authenticated engineer's project access.

bugsnag_errors_list

List errors

Fetch a paginated list of errors for a project, filtered by status, severity, or time range.

Parameters

Name

Type

Required

Description

project_id

string

Required

Project to list errors from

filters

string

Optional

Filter expression, e.g. error.status:open

sort

string

Optional

Sort order such as last_seen or events

bugsnag_error_get

Get error

bugsnag_events_list

List events

bugsnag_error_update

Update error

bugsnag_projects_list

List projects

bugsnag_release_get

Get release

Build your Agent

The same per-user auth pattern wires into every agent framework.

import { ScalekitClient } from "@scalekit-sdk/node";
import { ChatOpenAI } from "@langchain/openai";
import { createReactAgent } from "@langchain/langgraph/prebuilt";

const sk = new ScalekitClient(process.env.SCALEKIT_ENV_URL, process.env.SCALEKIT_CLIENT_ID, process.env.SCALEKIT_CLIENT_SECRET);

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["bugsnagmcp"], toolNames: ["bugsnag_errors_list", "bugsnag_error_get", "bugsnag_events_list"] },
  pageSize: 100,
});

const agent = createReactAgent({ llm: new ChatOpenAI({ model: "gpt-4o" }), tools });
await agent.invoke({ messages: [{ role: "user", content: "Triage open checkout errors" }] });

import OpenAI from "openai";
import { ScalekitClient } from "@scalekit-sdk/node";

const openai = new OpenAI();
const sk = new ScalekitClient(process.env.SCALEKIT_ENV_URL, process.env.SCALEKIT_CLIENT_ID, process.env.SCALEKIT_CLIENT_SECRET);

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["bugsnagmcp"], toolNames: ["bugsnag_errors_list", "bugsnag_error_get", "bugsnag_events_list"] },
  pageSize: 100,
});

const res = await openai.responses.create({
  model: "gpt-4o",
  input: "Show top open errors in checkout",
  tools,
});

import Anthropic from "@anthropic-ai/sdk";
import { ScalekitClient } from "@scalekit-sdk/node";

const anthropic = new Anthropic();
const sk = new ScalekitClient(process.env.SCALEKIT_ENV_URL, process.env.SCALEKIT_CLIENT_ID, process.env.SCALEKIT_CLIENT_SECRET);

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["bugsnagmcp"], toolNames: ["bugsnag_errors_list", "bugsnag_error_get", "bugsnag_events_list"] },
  pageSize: 100,
});

const msg = await anthropic.messages.create({
  model: "claude-sonnet-4-6",
  max_tokens: 1024,
  tools,
  messages: [{ role: "user", content: "Resolve the checkout TypeError" }],
});

import { LlmAgent } from "@google/adk/agents";
import { ScalekitClient } from "@scalekit-sdk/node";

const sk = new ScalekitClient(process.env.SCALEKIT_ENV_URL, process.env.SCALEKIT_CLIENT_ID, process.env.SCALEKIT_CLIENT_SECRET);

const { tools } = await sk.tools.listScopedTools("user_123", {
  filter: { connectionNames: ["bugsnagmcp"], toolNames: ["bugsnag_errors_list", "bugsnag_error_get", "bugsnag_events_list"] },
  pageSize: 100,
});

const agent = new LlmAgent({ name: "bugsnag_agent", model: "gemini-2.0-flash", tools });
await agent.run("List new errors in the latest release");

Try these prompts

Paste any prompt into your agent to start working Bugsnag errors.

Triage & resolve

Copy the prompt

Copied

What are the top open errors in the checkout project this week?

Copy the prompt

Copied

Mark the submit TypeError as fixed and reassign it to me.

Copy the prompt

Copied

Ignore the noisy NetworkError spike in the mobile project.

Investigate

Copy the prompt

Copied

Pull the latest stack trace for error 5f2a9c.

Copy the prompt

Copied

Show me events for the OutOfMemoryError over the last 24 hours.

Copy the prompt

Copied

Which devices are most affected by the checkout crash?

Report & recall

Copy the prompt

Copied

List all projects I have access to.

Copy the prompt

Copied

What is the stability score for the latest release?

Copy the prompt

Copied

How many events did the submit TypeError get this week?

SEE HOW AUTH WORKS

The engineer authorises Bugsnag once. Every later agent call uses their token with project scope enforced.

1

Authorize

Your user connects

Bugsnag MCP

once. We tie it to their identity and the meetings they approved — no shared bot account, no org-wide access

Who:

user ‘A’

when:

Once per user

access:

Limited to user

2

Store

Their

Bugsnag MCP

token lives in a vault scoped to them. User A's meetings are never reachable by an agent acting for user B, even on the same connection

vault:

encrypted

scope:

per-user

tokens:

auto-refreshed

3

Resolve

When your agent calls a

Bugsnag MCP

tool, we fetch the right token server-side. It never touches your agent, never appears in the LLM context, never shows up in your logs

speed:

~40ms

check:

before every call

seen by:

nobody

4

Audit

Every

Bugsnag MCP

tool call is logged — who triggered it, which meeting was fetched, what came back. 90 days of history, tied to the user who authorized it

history:

90 days

export:

SIEM-ready

logged:

every call

Test other agents

See the same per-user auth pattern across other monitoring connectors.

No items found.

Why Scalekit

Secure your agent's access. Connectors ship in minutes

Other connector libraries treat auth as a demo afterthought. Monitoring data needs real per-user attribution.

01.

Shared tokens break per-user analytics

A shared Bugsnag token looks fine in a demo. In production every fix and reassign logs as one service account, so you lose track of who acted. Scalekit resolves the real engineer's credential per call.

// shared token
fix → monitoring_bot

// scalekit
fix → user_abc ✓

02.

Authentication is not authorization

03.

Multi-tenancy is architectural

04.

Bugsnag today. Others tomorrow.

“Our agents act across Salesforce, Gong, Google Drive, and more, on behalf of every customer. Scalekit behind the scenes meant we can keep adding tools without ever rebuilding how credentials or tool calling work.”

Venu Madhav Kattagoni

Head of Engineering / Von

FAQs

Frequently Asked Questions

Does the agent access Bugsnag as the user or as a shared key?

As the user. Scalekit resolves the authenticated engineer's own Bugsnag credential at request time, so every error read, fix, and reassignment is attributed to them in Bugsnag's activity log rather than a shared monitoring bot.

Where is the Bugsnag token stored?

In an AES-256 encrypted vault, namespaced per tenant. Tokens are fetched only at the moment of a tool call, refreshed automatically before expiry, and revocable from a single dashboard, never exposed to the model or the prompt.

Can I limit what the agent is allowed to do in Bugsnag?

Yes. Pass a toolNames filter to listScopedTools to expose only the actions you want, for example read-only error listing without fix or reassign. Scalekit also runs a scope check before each API call.

What happens when a user revokes Bugsnag access?

The credential is invalidated at the next tool call, which fails closed rather than falling back to a shared token. Other users are unaffected, and the revocation is recorded in the audit trail.

Can the agent read release stability, or only work individual errors?

It can do both. Expose bugsnag_release_get to let the agent read stability scores, adoption, and new-error counts per release, so it can advise whether to ship or fix, while error write actions stay scoped per user to the projects the engineer already owns.

Bugsnag MCP

Tools your monitoring agent reaches for on Bugsnag, scoped per user.