> ## Documentation Index
> Fetch the complete documentation index at: https://docs.paperzilla.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# How do I use Paperzilla MCP with OpenClaw?

> Add Paperzilla to OpenClaw with Streamable HTTP and OAuth, then ask for public digests, paper Markdown, and your own project feeds.

> [!TIP]
>
> **AI agents**: This page is available as [markdown](https://docs.paperzilla.ai/answers/how-do-i-use-paperzilla-mcp-with-openclaw.md). See also the [docs index](https://docs.paperzilla.ai/llms.txt) and [full docs](https://docs.paperzilla.ai/llms-full.txt).

In OpenClaw, open **Settings → MCP → Add server**. Name it `paperzilla`, select **Streamable HTTP**, and enter `https://paperzilla.ai/api/mcp`.

After adding it, use the scoped MCP editor to set `auth: "oauth"` and `oauth.scope` to `catalog:read projects:read feedback:write markdown:generate offline_access`. Enable the server and select **Sign in**. Sign in or create a Paperzilla account, check the account and permissions, and select **Allow**.

Return to OpenClaw and ask: “Find public digests about robotics.” A Free account works, and you do not need a project for public digests or catalog papers. Personal feeds require an owned project. You can connect by URL without installing `pz` or a Paperzilla skill.

To rate or star a paper in your own project feed, ask explicitly. Feedback tools require `feedback:write`. For an existing connection, update the scopes and sign in again to approve that permission.

See [Use Paperzilla MCP with OpenClaw](https://docs.paperzilla.ai/guides/openclaw-mcp.md) for the full setup, permissions, example requests, connection checks, and troubleshooting. Manage or revoke access in [Connect AI clients](https://paperzilla.ai/ai-connections).


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.