Use Prompt Verse with your AI

Bring your own AI. Claude, ChatGPT, Grok or any assistant that supports MCP connects to your library, finds a prompt, asks for it rendered with its variables filled, and runs it on its own subscription. This page shows how to connect and what your AI can do once it is connected.

Your Prompt Verse address#

Prompt Verse has its own MCP server. It answers at POST /mcp on your Prompt Verse address, so the address ends in /mcp. You find yours in the app under Connect, with a button to copy it. It speaks MCP over HTTP and takes either a sign-in (OAuth) or an API key.

Every call acts in your own Prompt Verse workspace, after a plan check. Without a Prompt Verse plan, the call answers with PLAN_REQUIRED.

Connect your AI#

Claude#

In Claude, open Settings, then Connectors, then Add custom connector.
Paste your Prompt Verse address and save.
Sign in with your Prompt Verse account when Claude asks (OAuth: Claude registers itself and you approve it once).

ChatGPT#

In ChatGPT, open Settings, then Connectors, and add a remote MCP server.
Paste your Prompt Verse address.
Sign in with your Prompt Verse account when it asks.

Other MCP clients and scripts#

In the app, open Connect and make a key under Keys. It is shown once, so copy it then.
Point your client at your Prompt Verse address.
Send the key as a bearer token.

MCP client configuration

{
  "mcpServers": {
    "prompt-verse": {
      "url": "https://<your Prompt Verse address>/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Your first ask#

Paste this to your AI, with your own topic and input:

A first ask

Use Prompt Verse. Find my prompts about <topic> (prompts, action search), open the best one (prompts, action get), answer it for <my input>, and suggest one improvement I can save (prompts, action update).

For a prompt with variables, ask for the rendered text: prompts with action render, the prompt’s id, a format such as messages, and a variables object. The answer lists any variable you left out.

The tools#

Your AI sees a short list: one entry per area, with the detail on request. Ask for the actions of an area with action: "help", and for the exact arguments of one action with action: "help" and of.

Tool	What it is for	Actions
`prompts`	Prompts, responses and contexts of the workspace.	`create` `get` `list` `render` `run` `search` `share` `update`
`versions`	Git-shaped versioning of a prompt, workflow or agent.	`commit` `commits` `branch` `branches` `checkout` `current` `merge` `tag` `tags`
`evals`	Cases plus a scoring method, checked against a prompt.	`create` `get` `list` `run` `runs`
`agents`	Agents you own.	`create` `get` `list` `run`
`workflows`	Workflows you own.	`create` `get` `list` `update` `run`
`collections`	Named shelves of assets.	`create` `list` `get` `add` `remove` `update` `delete`
`library`	Pieces, skills, MCP tools and models.	`pieces` `piece` `skills` `skill` `create_skill` `tools` `create_tool` `call_tool` `models`
`runs`	Run history.	`list` `workflow`
`market`	The marketplace.	`browse` `read` `install` `buy` `publish` `withdraw` `try`

What your AI does instead#

The run actions of prompts, agents and workflows, evals run and market try are listed but answer AI_DISABLED, because your own AI does the running. The error tells your AI to call prompts render and to answer the text itself, which is the intended way. To test a prompt, your AI can read the saved cases with evals get, run the prompt and score the answers in the conversation.