Add Phersa to Gemini CLI

Find viral videos, make your own with your AI persona and run comment-to-DM, without leaving Gemini CLI. Gemini CLI connects to Phersa's remote MCP server over HTTP. Add it with one command, then sign in with /mcp auth.

In your terminal, run:

gemini mcp add --transport http -s user phersa https://mcp.phersa.com/mcp
Setup

Connect Gemini CLI to Phersa

Gemini CLI MCP Setup: Add Phersa.

phersa.com/gemini-cli
  1. Step 1: Add the server

    Run this in your terminal. -s user adds Phersa for all your projects; -s project adds it to the current one.

    gemini mcp add --transport http -s user phersa https://mcp.phersa.com/mcp
  2. Step 2: Sign in

    Start Gemini CLI, run the auth command, sign in in the browser and approve the access.

    /mcp auth phersa
  3. Step 3: Check the connection

    Type /mcp to list servers, then ask Gemini to run personas-list.

    /mcp

MCP server details

  • Endpoint: https://mcp.phersa.com/mcp · Streamable HTTP · OAuth 2.1 with PKCE. Metadata: https://mcp.phersa.com/.well-known/oauth-protected-resource/mcp. Dynamic client registration and client ID metadata documents are supported.
  • Permissions, safety and all 40 tools: phersa.com/mcp
Try it

What you can ask Gemini CLI

Sign in once and your agent can read and create. Nothing is generated until you confirm the price.

Gemini CLI · example prompts
  1. Read

    Find viral healthy drink reels, sort them by most viewed and show me the top five with their stats.

    Tools used: discover-search, video-get

  2. Read

    Break down the best one: the transcript, the hook and why it worked.

    Tools used: video-get

  3. Write

    Remake it with my AI influencer in a 9:16 project, write the image and video prompts for every clip, then show me the generation plan.

    Tools used: personas-list, create-project, create-clip, jobs-create (dry run)

In Gemini CLI

What is different about Phersa in Gemini CLI

  • gemini mcp add --transport http saves Phersa in settings.json with url and "type": "http". The older httpUrl key still works but is deprecated.
  • Gemini CLI is free with a personal Google account, up to 1,000 requests a day.
  • The Gemini app itself only accepts custom MCP servers through Gemini Spark, in the web app, for personal accounts aged 18+ in the US.
  • Gemini CLI finds Phersa's sign-in automatically and registers itself with dynamic client registration.

Gemini CLI questions

What is Gemini CLI?

Gemini CLI is Google's open-source AI agent that brings Gemini models into your terminal. It is free with a personal Google account and connects to local and remote MCP servers.

Gemini CLI on GitHub ↗

Can Gemini CLI make AI character videos?

Yes, with Phersa connected. Gemini CLI finds viral videos, plans the clips with your AI persona, shows you the price, and generates on your own API keys only after you agree. It edits and exports the video; you post it yourself, and it can set up comment-to-DM on that post.

Does Gemini support MCP servers?

Gemini CLI does: it connects to remote Streamable HTTP servers with OAuth and dynamic client registration. In the Gemini app, only Gemini Spark accepts custom MCP servers.

Check Gemini CLI MCP docs ↗

Is Gemini CLI free?

Yes. Gemini CLI is open source and free with a personal Google account, up to 1,000 requests a day.

Gemini CLI on GitHub ↗

How do I find viral videos in Phersa from Gemini CLI?

Ask for a topic, hook or phrase: Gemini CLI searches the library with discover-search and opens a video's stats, transcript and breakdown with video-get.

See example prompts →