Moogsoft MCP server: there isn't one
Moogsoft's job is to correlate alerts. No tool lets an agent ask Moogsoft which alerts it grouped.
Last verified 1 September 2026 · from GitHub search, the Moogsoft site and the official MCP registry
This page is one of 90 in a directory of monitoring and developer tools. Each page checks what an AI coding agent can get out of the tool through its MCP server, the connector that lets the agent query the tool directly.
Summary
Moogsoft has no MCP server. A GitHub search for repositories naming both Moogsoft and MCP returns zero results, and the official registry has no entry.
The site is still up: moogsoft.com returned HTTP 200 on 1 September 2026.
Moogsoft correlates a flood of alerts into a smaller set of situations, and an agent would want to read that result. Instead it has to work the grouping out again.
What is the Moogsoft MCP server, and is it usable?
Moogsoft is an AIOps platform. It ingests alerts from many monitoring sources and groups related ones into situations, so an on-call engineer sees one incident instead of two hundred alerts.
The product was absorbed into Dell APEX AIOps following acquisition, and neither the Moogsoft name nor the Dell offering publishes an MCP server.
For an agent, correlation is the valuable output. Raw alerts are available from whatever produced them; the grouping is the part only Moogsoft computed, and there is no tool that returns it.
Every one of the 90 vendors in this directory gets the same four checks. Moogsoft passes 0 of them.
| Check | Result | Evidence |
|---|---|---|
| Official | Fail | No MCP server. A GitHub search for repositories naming both Moogsoft and MCP returns zero results on 1 September 2026. |
| Maintained | Fail | There is nothing to maintain. The product moved into Dell APEX AIOps and no connector followed it. |
| Distributed | Fail | No package, and no Moogsoft entry in the official MCP registry. |
| Usable | Fail | The site still resolves, returning HTTP 200 on 1 September 2026, and carries no MCP documentation. |
Moogsoft works in production observability, and a search for repositories naming both Moogsoft and MCP returns zero results.
| Part of the work | What Moogsoft has here | Can an agent reach it |
|---|---|---|
| Code, tests & review | None | No code product, and no server |
| CI & release | None | No pipeline product, and no server |
| Production observability | Alert ingestion, correlation into situations, noise reduction | No. Alert correlation is the product, and no tool returns a situation |
| Agent observability & evals | None | No traces or evaluations, and no server |
| AI cost management | None | No spend data, and no server |
Moogsoft leaves an agent with raw alerts but not the grouping that identifies one underlying problem. The agent can query the monitoring tools that produced those alerts. It cannot ask Moogsoft for the correlation result.
What can an AI agent do with Moogsoft?
Moogsoft leaves the agent able to reach raw alerts but not the product's correlated situations.
Alerts are reachable from their sources; the correlation between them is not.
A GitHub search for repositories naming both Moogsoft and MCP returns zero results, and the official MCP registry has no entry, as of 1 September 2026.
The alerts Moogsoft ingests come from monitoring tools that often do publish servers. What is missing is the grouping Moogsoft computed on top of them.
Guard: Expect an agent to see raw alerts and to have to work out relationships itself, and do not assume it can read a situation.
Do you need the MCP server at all?
There is no Moogsoft server to need, so the question becomes which source to connect instead. For the alert side, connect the source. Prometheus exposes firing alerts and alerting rules through list_alerts and list_rules; Grafana, Datadog and Dynatrace all reach their own alerting surfaces.
For the incident side, which is where correlated alerts usually end up, PagerDuty publishes 101 tools, and its coverage includes alert grouping and event orchestration. This page names no individual PagerDuty tool for those two areas. incident.io documents 32 tools with alert_stats covering alert counts and workload from linked incidents, and Rootly exposes alert routing and grouping directly.
For correlation itself, incident.io's ask_telemetry and Rootly's find_related_incidents come closest: both do similarity analysis over history, which is a different technique from Moogsoft's but answers a similar question.
No MCP server in this directory reproduces Moogsoft-style alert correlation. An agent can read the inputs and reason about them, which is workable but is not the same as reading a computed situation.
Read alerts from their monitoring source, and use an incident platform's similarity tools where correlation matters.
Moogsoft MCP server setup
Nothing here installs a Moogsoft server, because there is none. This is the working route: read alerts from the platform that now handles them, and use its own grouping where correlation matters.
Context first. The product was absorbed into Dell APEX AIOps after the acquisition, so for most teams the practical question is what the replacement exposes rather than what Moogsoft does.
Three incident platforms in this directory publish servers, and each takes a different shape of credential.
PagerDuty, read-only, in Claude Code
PagerDuty publishes a first-party server, and a user API key scoped to reads keeps an agent to querying incidents rather than acting on them.
The key is PagerDuty's own, and the permissions attached to it in PagerDuty decide what returns.
claude mcp add-json "pagerduty" '{"command":"uvx",
"args":["pagerduty-mcp"],
"env":{"PAGERDUTY_USER_API_KEY":"<your-key>"}}'incident.io, hosted
incident.io runs a hosted server and documents 32 tools across incidents, alerts, escalations, schedules, the catalog and follow-ups.
No credential goes in the configuration. The endpoint authorizes in the browser on first use, so nothing is stored in the file.
claude mcp add --transport http incidentio \ https://mcp.incident.io/mcp
Rootly
Rootly publishes a first-party server on PyPI. It is the third option here because its shape differs: a local process holding an API token.
The token is Rootly's own, and its scope in Rootly bounds the answer.
claude mcp add-json "rootly" '{"command":"uvx",
"args":["rootly-mcp-server"],
"env":{"ROOTLY_API_TOKEN":"<your-token>"}}'Every other client
Each block below is the configuration for one client, with the file path and the key that client expects.
What none of them replaces
Moogsoft's product was the correlation: taking a flood of alerts and reducing it to a smaller set of situations. That reduction is the judgment no connector returns.
Incident platforms group alerts too, so the gap is narrower than it looks, but the grouping is theirs and not Moogsoft's. This is a comparison and involves no credentials.
Moogsoft : many alerts -> few situations (not exposed) Replacement: many alerts -> its own grouping
Read-only access, permissions and security
No Moogsoft credential is created
With no server and the product absorbed elsewhere, nothing here asks for a Moogsoft token.
Check what your contract now covers
The product moved into Dell APEX AIOps, so an integration planned against the Moogsoft name may not describe what you actually have.
Incident data is unusually sensitive
Alerts, timelines and postmortems name people, systems and failures, and every connected tool sends that to a model.
Prefer a read-scoped key
PagerDuty and Rootly both take a token whose permissions decide how much the agent can change, so scope it to reads before connecting.
The hosted route stores nothing locally
incident.io authorizes in the browser, which keeps a long-lived token out of your configuration file entirely.
Troubleshooting
- You cannot find a Moogsoft MCP server
- There is not one. A GitHub search for repositories naming both Moogsoft and MCP returned zero results on 1 September 2026.
- The site is live, so the product should still be separately supported
moogsoft.comreturned HTTP 200 on 1 September 2026, so the domain is not dead. The product itself was absorbed into Dell APEX AIOps after the acquisition.- Check what your contract actually covers before planning an integration against the Moogsoft name.
- You want correlated situations rather than raw alerts
- That correlation is exactly what no tool returns. Moogsoft reduces a flood of alerts into a smaller set of situations, and an agent cannot read that result.
- Connect the alerting source instead and accept that the agent sees the flood rather than the summary.
- A registry search returns nothing
- That matches. The official MCP registry has no Moogsoft entry and no package exists.
Moogsoft MCP server: Reference
| Item | Value |
|---|---|
| Server published | None |
| Repositories naming Moogsoft and MCP | 0 |
moogsoft.com HTTP status | 200 on 1 September 2026 |
| MCP documentation on the site | None |
| Product status | Absorbed into Dell APEX AIOps after acquisition |
| Packages published | None |
| Official MCP registry results | None |
| What is unreachable | The correlation: which alerts Moogsoft grouped into one situation |
| Checked | 1 September 2026 |
What engineers report
The site being live is the useful nuance here, because it separates a missing feature from a discontinued product.
| What was checked | What it shows |
|---|---|
| Repositories naming Moogsoft and MCP | 0 |
moogsoft.com HTTP status | 200 on 1 September 2026 |
| Official MCP registry results | None |
| Product status | Absorbed into Dell APEX AIOps after acquisition |
GitHub was searched for repositories naming both Moogsoft and MCP on 1 September 2026, returning zero results. The Moogsoft site returned HTTP 200 on the same date. The official MCP registry has no Moogsoft entry.
Should you connect an agent to Moogsoft?
No. There is nothing to connect. Best for a monitoring connector for the alerts, plus an incident platform for the grouping.
- Strongest fit
- Teams already running Moogsoft who will read alerts from their sources instead.
- Main advantage
- Three incident platforms in this directory publish servers, and each groups alerts in its own way.
- Main weakness
- Alert correlation, the product's entire output, cannot be queried by an agent.
- Operational risk
- Planning an integration against the Moogsoft name when the product moved into Dell APEX AIOps after the acquisition.
From Oodle
One platform for agent traces and infrastructure
Agent traces usually sit in a different product from the rest of your telemetry, so when a slow database makes an agent slow the symptom is in one tool and the cause is in another. Oodle keeps both in one query surface, from $10 per million spans.
See agent observabilityFrequently asked questions
Does Moogsoft have an MCP server?
No. A GitHub search for repositories naming both Moogsoft and MCP returns zero results, and the official registry has no entry, as of 1 September 2026.
Is Moogsoft still a product?
The site returned HTTP 200 on 1 September 2026. The technology was absorbed into Dell APEX AIOps after acquisition, and neither name publishes a server.
Can an agent read the correlated situations?
No. It can read the underlying alerts from whichever monitoring tool produced them, but the grouping Moogsoft computed is not exposed.
What comes closest?
Rootly's find_related_incidents and incident.io's ask_telemetry both do similarity analysis over incident history, which answers a similar question by a different method.