Getting started | RankControl Docs

Install the RankControl CLI, authorize it against your workspace, and run your first commands.

RankControl publishes ranking content to your own site and tracks whether AI engines cite you. This documentation covers the three developer surfaces: the CLI, the MCP server for AI agents, and the REST API they both sit on.

Install

npm install -g rankcontrol

The package ships two binaries, rankcontrol and the shorter rctrl. One-off use without installing works too: npx rankcontrol <command>.

Authorize

Start the login

rankcontrol login

The CLI prints a confirmation code and opens your browser.

Approve in the dashboard

The approval page shows the same code and the exact permissions being granted. Check the codes match, then approve. The key is scoped to the workspace you are signed into.

Done

A scoped API key is saved to ~/.rankcontrol/config.json (file mode 0600). The key appears in Settings → API in your dashboard, named after your machine, and can be revoked there at any time.

Prefer explicit keys? Create one in Settings → API and export it instead:

export RANKCONTROL_API_KEY=rctrl_pk_...

First commands

rankcontrol funnel     # AI pipeline, last 30 days: crawls, AI visits, AI leads
rankcontrol ideas      # scored content-idea backlog
rankcontrol report     # executive summary, last 30 days vs the 30 before
rankcontrol --help     # everything

Every command prints JSON, so output pipes cleanly into jq or any script.

Default scopes

rankcontrol login requests read access to citations, content, leads, and analytics, plus write:content and write:publish. Pass --scopes read:content,write:content to request less. Team, backlink, and social write scopes are opt-in.

Where next

CLI reference

Every command, grouped by area, with flags and examples.

MCP server

Give Claude, Cursor, or any MCP client your workspace as tools.

REST API

Endpoints, scopes, rate limits, and the dry-run pattern.