When a monitoring check fails, the useful next step is often to inspect the result and understand what changed. SolidPing brings that work into a coding-agent session with a plugin for Codex and Claude Code that connects to the SolidPing remote MCP server.
SolidPing is an open-source uptime-monitoring project that can be self-hosted. Its project README currently describes 44 check types, multi-region workers, private agents, status pages, incidents, and on-call escalation in a single Go binary. The project is licensed under AGPL-3.0; the separate plugin repository uses the MIT license.
Read monitoring context, then act with confirmation
Through the plugin, a developer can ask an agent to list checks or incidents, inspect results, and diagnose a failing check. The available tools also include creating checks and status pages. The plugin repository says tools are not pre-approved: the agent asks before each call unless the user has allowed it. It also notes that no code runs on the user's machine; the plugin connects to the remote service through MCP and OAuth sign-in.
For example, a developer can ask which checks are down, investigate why an API health check is failing, or review incidents from the past week. If a monitoring change is needed, the agent can prepare the corresponding write action for confirmation.
Hosted and self-hosted paths
The project offers a hosted service with a free tier, while teams that prefer to operate their own instance can configure the MCP endpoint in the plugin or register the server directly. The repository documents installation for both Codex and Claude Code.
Explore the SolidPing plugin, the SolidPing project, its website, documentation, and the demo.
