An ops AI Agent that understands your infrastructure, finds the root cause, and fixes it — right from Slack, Telegram, Lark or DingTalk.
Ongrid is an operations AI agent that investigates infrastructure problems by correlating metrics, logs, traces, topology, and alerts, then can carry out remote actions. SRE and DevOps teams use it through Slack, Telegram, Lark, DingTalk, and other supported chat channels to diagnose systems and manage Kubernetes resources. The catalogue entries provide agents, skills, and instructions for operating Ongrid.
This repository also configures its own agents. See what ongrid tells them →
Skill Claude CodeCodex
An Ongrid MCP skill for investigating a selected alert using the configured monitoring service. Ongrid is a monitoring system, and an alert is a notice that a device or measurement may need attention.
Skill Claude CodeCodex
An Ongrid MCP skill for checking a specified host through the configured monitoring service. A host is a computer or other networked system, and load describes how busy its resources are.
Skill Claude CodeCodex
A tool for producing inspection reports from Ongrid data for a specified object and time range. The report lists abnormalities, supporting evidence, and areas not covered.
Skill Claude CodeCodex
A tool for carrying out explicitly requested Ongrid operations involving devices, alerts, rules, dashboards, platform APIs, restricted hosts, or Kubernetes.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: