Niches · Observability & incidents

Triage production errors and turn them into a code fix

One company's docs show a full fit.

1companies, full fit
3companies, fit with a limit
13companies carry the tag
20share of the category's work
10first-hand accounts

What counts as done

For each chosen issue: the full stack trace, affected-user count and first and last seen time retrieved, the source file located, and either a pull request or a written reason for not fixing it.

How agents ask for it: fix the top Sentry issues, what is this error and where in the code, open a PR for this crash, which errors affect the most users

Does the job 1

The docs show it meets the test above, and nothing we read says otherwise.

ListingWhy
Sentry MCP B
Functional Software, Inc. d/b/a Sentry
get_issue_details, get_event_stacktrace (full trace per thread), get_issue_breadcrumbs and find_releases cover the issue; acc-02 runs top-issues-by-users to a pull request. Whether resolved source links are now in the payload (#1157, closed) wasn't confirmed. source

Does the job, with a limit 3

It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.

ListingWhy
Better Stack C
Better Stack, Inc.
List errors gives affected users and state, Get error details gives call site and first occurrence, releases give first and last seen; the full trace needs an exceptions query, and resolved source links are not documented. source
Datadog MCP Server D
Datadog, Inc.
The error-tracking toolset searches and reads Error Tracking Issues and counts errors by version in SQL, but the docs do not show whether the issue read carries the full stack trace, affected users or source links. source
New Relic MCP Server B
New Relic, Inc.
list_entity_error_groups and get_error_statistics give error groups with counts and first and last seen, and search_errors returns raw events, but the docs don't say whether full stack frames, affected users or source links come back. source

Couldn't check 1

The docs were closed to us, or silent. These carry no mark either way.

ListingWhy
Cronitor MCP server C
Cronitor
list_site_errors and get_site_error exist, but the docs leave their fields to tools/list behind sign-in, so stack depth, affected users and first and last seen could not be read.

Doesn't do this job 7

Listings in the category that do something else, with the reason.

ListingWhy
Axiom MCP Server BB
Axiom, Inc.
No error-tracking issue object: no grouped issue, affected-user count, first and last seen or source-linked frames. Stack traces exist only if the app logged them, and the agent would compute the rest in APL. source
Causely MCP server E
Causely, Inc.
Diagnoses list exceptions as evidence, but no tool returns a full stack trace, affected-user count or first and last seen time per error. source
Grafana MCP Server BB
Raintank Inc., d/b/a Grafana Labs
No error-tracking issue object: no grouped errors, affected-user counts, first and last seen or source-linked frames. source
Honeycomb MCP B
Hound Technology, Inc. d/b/a Honeycomb
No error-tracking issue object: no grouped errors, affected-user count, first and last seen or source-linked frames. source
incident.io API B
Pineapple Technology Ltd.
No error-tracking data: no stack traces, affected-user counts or first and last seen per error. source
PagerDuty MCP Server E
PagerDuty, Inc.
No error-tracking data or stack traces. source
Rootly MCP Server C
Rootly Inc.
No error-tracking data or stack traces. source

What stops agents finishing

  • stack trace or code-mapping link missing from the API payload
  • OAuth-only hosted endpoint with no headless route
  • single dispatcher tool hides writes from host approval

Around it

Before: observability.scheduled-watch, notifications.alert-received. After: code.open-pr, project-management.create-ticket.

Where this comes from

Weight 20 of 100 in Observability & incidents: 10 of 42 accounts; Sentry dominates (9 of 10), so this weight describes Sentry users more than the category

The same as data: /api/v1/use-cases/observability.error-to-fix.json, or the MCP tool get_use_case with id observability.error-to-fix. Other jobs in this category: /api/v1/niches?category=observability.

Draft, read 2026-11-15 from public docs and first-hand accounts, not yet reviewed by a person. A mark says what the docs show, not what a run measured.

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.