Is Glen a runbook tool or a monitoring system?

No. Glen is shared memory for the AI agents you already use. It captures the operational knowledge, root causes, quirks, and fixes, that monitoring and wikis miss, and feeds it back to agents in real time.

Does the whole on-call rotation share the same memory?

Yes. Glen is org-scoped, so every engineer's agent and every MCP client reads and writes one shared operational memory, regardless of who is paged.

Shared memory for DevOps engineers

DevOps runs on operational memory: which runbook actually fixed the last outage, why a service has that one weird flag, the order you have to restart things in, the deploy that broke prod and the rollback that saved it. Most of that lives in someone's head, a stale wiki, or a buried incident channel. When you bring AI agents into the loop, to triage alerts, draft runbooks, or reason about infrastructure, they start blind to all of it and rediscover the same operational quirks every time. Glen, shared memory for AI agents, gives the agents your team uses long-term shared memory as a single MCP tool, so operational knowledge persists and is shared across the on-call rotation.

Wire your AI tools to Glen over MCP and any agent, triaging an incident, reasoning about a deployment, or drafting a change, can call one tool that retrieves relevant long-term context and records new facts in a single round trip. Before it acts, it pulls what the team already knows: the fix from the last similar incident, the dependency that always trips people up, the maintenance window that matters. As the work happens, it writes back what was learned, a new failure mode, a corrected runbook step, the real root cause, so the next on-call engineer's agent starts informed.

The value is shared operational memory that survives rotations. Glen is org-scoped, so the knowledge is shared across every DevOps engineer, every agent, and every tool in your organization, not trapped with whoever happened to be paged. A lesson from one incident is available to everyone next time, and because Glen is a standard MCP server, the same memory is readable from Claude Code, Cursor, your incident-response agent, or any other MCP client. The tribal knowledge that usually walks out the door becomes durable memory your agents act on. You connect once over OAuth or an API key and stop relearning your own infrastructure during every incident.

FAQ

Is Glen a runbook tool or a monitoring system?: No. Glen is shared memory for the AI agents you already use. It captures the operational knowledge, root causes, quirks, and fixes, that monitoring and wikis miss, and feeds it back to agents in real time.
Does the whole on-call rotation share the same memory?: Yes. Glen is org-scoped, so every engineer's agent and every MCP client reads and writes one shared operational memory, regardless of who is paged.