Gemini CLI is Google’s open-source, terminal-based AI agent. It speaks the Model Context Protocol, so you can wire in ContextBolt SEO and pull live keyword, SERP, domain, and competitor data straight into your terminal workflow.
Every lookup you run lands on your SEO Dashboard, a private page you can sign in to from any machine, instead of disappearing into the session.
SEO research where you ship
When you are building, the question “what should this rank for, and who already owns it?” is one you want answered without breaking flow. With ContextBolt SEO in Gemini CLI you ask it in the terminal and the numbers come back inline. Every answer is saved to your SEO Dashboard, with no export and no second tool.
Because these are MCP tools, Gemini CLI chains them. “Find ten low-difficulty keywords for my docs site, check which ones a competitor already ranks for, and list the gaps” is one prompt, not ten dashboard sessions, and the whole session lands on your dashboard.
The config entry
Add ContextBolt SEO to ~/.gemini/settings.json (global) or .gemini/settings.json (per project), under the mcpServers object:
{
"mcpServers": {
"contextbolt-seo": {
"httpUrl": "https://seo.contextbolt.app/mcp"
}
}
}
Note the key is httpUrl, not url. Gemini CLI uses httpUrl for a remote streamable-HTTP server, reserves url for SSE endpoints, and uses command plus args for local stdio servers. This is the one detail that trips people up, since Cursor and Codex use url and Windsurf and Antigravity use serverUrl. Save the file, start Gemini CLI, and run /mcp auth contextbolt-seo. Enter your SEO account email and email code, then allow access. Run /mcp to confirm the tools loaded. The shared URL contains no personal secret. The full tour of tools and workflows is in the ContextBolt SEO guide.
Claude Code
Codex
Muse Code
Cursor
ChatGPT
Windsurf
Gemini CLI
Antigravity
VS Code
Cline
n8n