# Quillm [Health: Active]

**Category:** 💻 Developer Tools  
**Repository:** https://github.com/streamfog/quillm-plugin  
**GitHub Stars:** 0  
**Views:** 0  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/quillm

## Description
Dashboards, trackers and docs your agents build and keep current. Your team reads them by link.

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "quillm": {
    "url": "https://quillm.ai/mcp"
  }
}
```

## Documentation

## What the Quillm MCP server does

The Quillm MCP server gives agents access to Quillm workspaces containing shareable pages and separate datasets. It is intended for small internal tools such as dashboards, trackers, decision documents, and calculators. Each page has its own address, so people on the team can open it as a document without signing into an AI product.

Agents can inspect workspace information, read and modify datasets, create pages, update existing pages, and ask Quillm to test-render a page before it is shown to readers. They can also share a page by link when requested, add notes for a later agent, and submit a problem report about Quillm.

Data is kept apart from the page itself. That lets a recurring agent add new rows for each run without regenerating the page layout. Quillm retains revisions, and changes are signed by the agent that made them.

## How it works

The repository provides the connection details for Quillm's hosted MCP service at `https://quillm.ai/mcp`, along with an agent skill describing practices for maintaining pages over time. The hosted server is operated by Streamfog and is not included in this repository.

When an agent first uses a tool, the normal sign-in flow opens a browser. The user signs in to Quillm with Google, selects an existing workspace or creates one, and chooses whether the agent can only read or may also make changes. Workspace administrators can later restrict the agent to selected collections, sign it out, or revoke access. The workspace's Agents page shows connected agents.

## Setup and configuration

To connect the Quillm MCP server, use the setup method for the client you run. Claude Code can install the plugin from the `streamfog/quillm-plugin` marketplace and authenticate through `/mcp`. Claude Desktop, claude.ai, and Cowork can use the connector directory or the custom endpoint. Cursor, ChatGPT, Codex, and Grok Build have their own marketplace or plugin-directory entries described by the project.

Other MCP-compatible clients can add `https://quillm.ai/mcp` as a remote Streamable HTTP server. Client-specific instructions are available at `https://quillm.ai/connect`.

For CI jobs, cron tasks, or machines without a browser, a workspace administrator can create an agent token and pass it as an HTTP Bearer header. For example:

```sh
claude mcp add --transport http quillm https://quillm.ai/mcp --header "Authorization: Bearer qlm_…"
```

Store that token in the job's secret manager rather than in source control.

## Tools and capabilities

The available operations cover:

- Reading workspace information.
- Reading and writing datasets.
- Creating and updating pages.
- Test-rendering pages before publication.
- Sharing pages by link when requested.
- Leaving notes for a subsequent agent.
- Reporting a Quillm problem.

The project links to a complete tool list that describes what each operation reads or changes and its context usage.

## Limitations and notes

The Quillm MCP server depends on the hosted Quillm service; the repository does not contain the server implementation. Access is controlled at the workspace level, with either interactive OAuth sign-in or an administrator-created token. The material provided does not specify offline operation or local hosting.

Quillm's privacy policy documents stored data and information sent to other services. Workspace administrators can export the workspace contents. The repository is licensed under MIT.

_Full upstream README: https://allmcps.com/mcp/quillm/readme_