Built so an agent can run the day without paging you.
Most observability APIs were designed for humans clicking through a dashboard. The retries get wrapped in an SDK; the auth gets hidden behind an OAuth dance; the errors arrive as 500-page release notes. None of that ports to an agent.
TL;DR
Most observability APIs were designed for humans clicking through a dashboard. The retries get wrapped in an SDK; the auth gets hidden behind an OAuth dance; the errors arrive as 500-page release notes. None of that ports to an a...
Built so an agent can run the day without paging you.
Most observability APIs were designed for humans clicking through a dashboard. The retries get wrapped in an SDK; the auth gets hidden behind an OAuth dance; the errors arrive as 500-page release notes. None of that ports to an agent.
We built this surface for the case where the caller is a script you wrote at midnight, or a Claude/GPT/Gemini tool call that has to make a decision in the next 300 ms. Below is the short list of what that actually means — narrated against the live API.
Plug 24observe into any MCP client — no integration code.
Beyond the tool-definition JSON below, we ship a Model Context Protocol server (@observe24/mcp).
Add it to Claude Desktop, Claude Code, Cursor, or your own agent with one config block and your API token, then ask
things like "investigate incident 159" or "open a case and attach the brute-force incident." It's a thin,
safe proxy — every read plus a curated allow-list of safe writes, never DELETE, never auth
or billing — and it runs against your own self-hosted instance too.
// claude_desktop_config.json (or Cursor mcp.json)
{
"mcpServers": {
"24observe": {
"command": "npx",
"args": ["-y", "@observe24/mcp"],
"env": { "OBSERVE24_TOKEN": "obs_your_token" }
}
}
}
Available today by self-hosting from source — build it and set "command": "node",
"args": ["/path/to/observe24/apps/mcp/dist/index.js"]. The managed
@observe24/mcp package shown above publishes shortly.