MCP
helloLOG exposes a Model Context Protocol server so you can ask an AI tool about your own activity log instead of digging through the dashboard by hand — "did anyone touch checkout settings this week", "who published that page", "show me failed logins from the last 24 hours".
The server is scoped to one site: it authenticates with the same site key you use for the read API, and it can only ever see that site's events and stats.
Connect
URL
https://api.hellolog.io/v1/wordpress-activity-audit-log/mcp
Header
Authorization: Bearer <your site key>
Claude Code
claude mcp add --transport http hellolog \
https://api.hellolog.io/v1/wordpress-activity-audit-log/mcp \
--header "Authorization: Bearer <your site key>"
Claude Desktop
Add a remote HTTP server with the same URL and header from Claude Desktop's connector settings.
What it can see
The server answers questions grounded in your site's own event stream — recent activity, counts and breakdowns, and lookups by user, object, or event type. It's read-only: nothing an AI tool does through MCP can change your site or your logs.
Revoking access
MCP access is tied to your site key, so it follows the same rules as the read API: rotate or revoke the key from the dashboard and the connection stops working immediately — no separate step needed.