How to Connect Rollbar to Microsoft Copilot via MCP

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → Rollbar, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

Rollbar + Microsoft Copilot

Rollbar error monitoring in any AI agent: items with their history and occurrences, full stack traces, top active items and occurrence reports, the Metrics API, deploys, environments and code versions, RQL queries, and the account's projects, users and teams. It also resolves, mutes, assigns and snoozes items, comments on them and reports deploys.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

ROLLBAR_ACCESS_TOKEN

Step 1 — Get credentials

Setting up takes about 3 minutes. You need a Rollbar account and one access token. Which kind of token you paste decides what the connector can reach:

1. One project: in the project open Settings, Project Access Tokens and create or copy a project access token. Give it the read scope for reading, write to change items and post_server_item to report deploys. A project token always works on its own project, but it cannot list projects, users or teams.
2. The whole account: create an account access token instead. A service token (account Settings, API Access Tokens, Service tab) reaches every project; a personal access token (User Settings, Personal access tokens) can do only what its user can do. Their scopes are read and write. With an account token the AI passes project_id (from rollbar_list_projects) on project tools, and rollbar_list_items searches every project the token can read unless you name project_ids.
3. Paste the token into ROLLBAR_ACCESS_TOKEN. It is sent as the X-Rollbar-Access-Token header to https://api.rollbar.com/api/1.
4. Install the connector and run rollbar_list_items to check the token.

Only a personal access token can comment on items, and only a project token with post_server_item can report deploys.

What is broken right now: rollbar_get_top_active_items shows the items with the most occurrences in the last hours (1 to 168, default 24) with hourly counts, like the Top 10 on the project dashboard. rollbar_list_items lists and searches items (grouped errors and messages) by status, level, environment, framework, assignee, snooze state or the same search text as the Items page. rollbar_get_item reads one item. The number in a Rollbar URL (#123) is the project counter, not the item id: rollbar_get_item_by_counter turns a counter, or the UUID of an occurrence, into the item.

Investigating: rollbar_list_item_occurrences lists the single events of one item and rollbar_list_occurrences the latest events of the whole project, each with timestamp, code version and payload. rollbar_get_occurrence returns the full payload as in the Raw JSON view: exception class and message, stack trace frames, request, person, server, client, custom data and telemetry. rollbar_list_item_events shows who commented on, changed, assigned, snoozed or merged an item.

Triage: rollbar_update_item resolves an item (optionally with resolved_in_version, for example a git SHA), mutes or reactivates it, changes its level or title, assigns it to a user or team (ids from rollbar_list_users and rollbar_list_teams) or snoozes it for a number of seconds (snoozing needs a paid plan). rollbar_add_item_comment adds a comment in Markdown to the item's timeline. These need the write scope.

Trends and metrics: rollbar_get_occurrence_counts gives occurrences per minute, hour or day, rollbar_get_activated_item_counts new and reactivated items per day, and rollbar_get_item_metrics occurrence counts with first and last occurrence for a list of items. On the Advanced plan, rollbar_query_occurrence_metrics queries the Metrics API with filters, grouping and aggregates (for example occurrences per environment per day, or the top browsers of an item), and rollbar_get_resolution_time_metrics gives the time to resolve items (median, mean, min or max) per period.

Releases: rollbar_list_deploys, rollbar_get_deploy and rollbar_list_environments read deploys and environments. rollbar_report_deploy records a deploy of a revision to an environment, and rollbar_update_deploy finishes one reported as started (succeeded, failed or timed_out). To see what a release broke, rollbar_get_version counts new, reactivated, repeated and resolved items of one code version in one environment, and rollbar_list_version_items with event: new lists the items first seen in it.

RQL: for ad hoc analysis the AI starts a query in Rollbar's SQL-like language with rollbar_create_rql_job (tables item_occurrence and deploy), polls rollbar_get_rql_job until status is success and reads the rows with rollbar_get_rql_job_result. rollbar_list_rql_jobs and rollbar_cancel_rql_job manage earlier and running jobs. RQL changes no data.

Account: with an account token, rollbar_list_projects, rollbar_get_project, rollbar_list_users, rollbar_get_user and rollbar_list_teams.

Switched off at install: rollbar_request_person_deletion, which erases a tracked person (the end user attached to occurrences) from every project of the account, for example for a GDPR request. It cannot be undone, so a workspace admin turns it on when needed; rollbar_get_person_deletion_status then follows the job.

Data formats: every answer is {"err": 0, "result": ...}; err: 1 comes with a message. rollbar_report_deploy answers {"data": {"deploy_id": ...}}. All timestamps are Unix seconds in GMT, and report graphs are [timestamp, count] pairs. Item status is active, resolved, muted or archived; level is debug, info, warning, error or critical. Occurrence lists leave out request headers, cookies and notifier details; the full payload of rollbar_get_occurrence may contain personal data of your end users.

Paging: items come 100 per page; occurrences, deploys, environments and item events 20 per page by default, with limit up to 5000 where offered. Pass page from 1; occurrence lists can also page with last_id. total_count is null on projects with more than 100,000 items.

Limits: each token has its own Rollbar rate limit; X-Rate-Limit-Remaining shows what is left. The Metrics API tools need the Advanced plan.

Common errors: 401 means a missing or malformed token. 403 means the token is disabled, lacks the scope, is a project token on an account tool, or (Requires project access) is an account token without a usable project_id. 404 is an unknown id or one of another account, 422 invalid parameters and 429 the token's rate limit.

The connector does not send errors to Rollbar (that is the job of the Rollbar SDK in your app), does not delete items, and does not create projects, teams, users or access tokens.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in Microsoft Copilot

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

Microsoft Copilot uses MCP servers through agents built in Copilot Studio: you add AnythingMCP to an agent as a tool, then publish the agent to Microsoft 365 Copilot and Teams. Copilot Studio connects from Microsoft's cloud over Streamable HTTP, which AnythingMCP supports.

1. In Copilot Studio, open your agent (or create one), go to Tools and choose Add a tool → New tool → Model Context Protocol.
2. Enter a Server name (for example AnythingMCP), a Server description that says what the connector does, and the Server URL https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID. Copilot Studio uses the description to decide when to call the server.
3. Under authentication, choose API key, set Type to Header and the header name to X-API-Key, then select Create.
4. In Add tool, choose Create a new connection, paste an MCP API key from AnythingMCP (MCP Servers → your server → API keys) and select Add to agent.
5. Try it in the agent's Test pane: ask for something that needs the Rollbar tools.
6. Select Publish, then Channels → Microsoft 365 and Microsoft Teams, keep Make agent available in Microsoft Copilot selected and choose Add channel. See agent in Teams installs it for you; in Microsoft 365 Copilot, type @ and pick the agent.

Prefer per-user sign-in? Choose OAuth 2.0 → Dynamic discovery instead of an API key: each user then signs in to AnythingMCP the first time the agent calls it.

Self-hosting? Use your instance's public HTTPS address instead of https://cloud.anythingmcp.com. Copilot Studio connects from Microsoft's cloud, so localhost does not work.

Available tools

Example prompts

• "What are the top errors in production over the last 24 hours, and which ones are new?"
• "Show me the full stack trace of the latest occurrence of Rollbar item #482."
• "Which items were first seen in version 3f9c2ab in production? Group them by level."
• "Resolve the TimeoutError item in version 1.8.3 and assign the NullPointerException to the backend team."
• "How many error and critical occurrences did we have per day this week compared with last week?"

FAQ

Which Copilot is this for? Microsoft 365 Copilot at work, through agents built in Copilot Studio. The Microsoft Copilot app for personal accounts has no option to add your own MCP server.

What do I need? A work account with access to Copilot Studio (a license or a trial) to build the agent. People who use it in Microsoft 365 Copilot or Teams need access to the agent: share it with them, or ask an admin to publish it to the organization. If your organization applies data policies to Power Platform connectors, they cover MCP servers too.

The agent doesn't call the tools. Check that the header name is exactly X-API-Key and that the key belongs to the same MCP server as the URL, and make the server description specific: Copilot Studio uses it to decide when to call the server. Publish the agent again after each change; in an open chat, type "start over" to get the new version.

Next steps

• Connector landing page
• Connect to Claude
• Connect to ChatGPT
• Connect to Cursor
• Connect to Gemini
• Connect to Meta Muse