Quillm

ai.quillmv1.0.0Updated Oct 3, 2026

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

VerifiedStreamable HTTPWeb executableData & AnalyticsProductivity & WorkflowKnowledge & Memory

Overview

AI-generated overview

Lets an assistant build and keep current shared dashboards, trackers, decision docs and calculators that teammates open by link.

What it does
Quillm stores the small tools an agent builds for a team: dashboards, trackers, decision docs and calculators. Each page lives at one address teammates open like a doc, with no AI account needed. Numbers live in datasets separate from the page, so an agent keeps a page current by adding rows run after run instead of rewriting it. The agent can read the workspace, read and write datasets, create and update pages, test-render a page, share a page by link when asked, leave notes for the next agent, and report a problem.
When to use it
Use it when a team wants agent-produced pages that stay useful over time, such as recurring metric trackers or decision docs read by link. It suits scheduled agents that append each run's numbers to a dataset without touching the page. It is not for one-off answers that never need to be shared or updated.
Requirements
A remote Streamable HTTP endpoint at quillm.ai/mcp; no local runtime or package. Sign-in uses OAuth in a browser with Google, choosing or naming a workspace and deciding whether the agent may change things or only read. For headless use (CI, cron, servers), an admin creates an agent token on the workspace's Agents page and passes it as an Authorization Bearer header; the token belongs in a secret store, not a repository.
Before you install
The agent can write: it reads and writes datasets, creates and updates pages, and can share a page by link when asked, so pages may become visible outside the workspace. Access is granted at sign-in, where you choose read-only or change permissions, and admins can limit an agent to some collections, sign an app out, or revoke it. The Authorization Bearer token is a credential and should be kept in a secret store. What Quillm stores and sends to other services is covered by its privacy policy.

Installation

In SourceWeft

  1. Open Quillm in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "quillm": {
      "type": "http",
      "url": "https://quillm.ai/mcp"
    }
  }
}

README

[Image]

Quillm plugin

Quillm keeps the small tools your agents build for your team: dashboards, trackers, decision docs and calculators. Each page lives at one address your team opens like a doc, with no AI account needed. The numbers live in datasets apart from the page, so an agent keeps a page current by adding rows, run after run, without rewriting it. Every save is test-rendered before readers see it, every revision is kept, and every change is signed by the agent that made it.

This repository connects your agent to Quillm. It holds:

  • the connection to Quillm's hosted MCP server, https://quillm.ai/mcp, which you sign in to with OAuth in your browser;
  • one skill, skills/quillm/SKILL.md, that teaches the agent the habits that keep pages useful for months, including how a scheduled agent should add each run's numbers without touching the page.

The server itself is hosted by Streamfog and is not part of this repository.

Install

AppHow
Claude Codeclaude plugin marketplace add streamfog/quillm-plugin, then claude plugin install quillm@quillm. Run /mcp and choose Authenticate.
claude.ai, Claude Desktop, CoworkSettings → Connectors → find Quillm in the directory, or add a custom connector with https://quillm.ai/mcp.
ChatGPT, CodexFind Quillm in the plugin directory.
CursorInstall Quillm from the Cursor Marketplace.
Grok BuildInstall Quillm from the plugin marketplace.
Anything else that speaks MCPAdd https://quillm.ai/mcp as a remote (Streamable HTTP) server and sign in.

Step-by-step instructions for each app: https://quillm.ai/connect

The first time a tool is used, a browser window asks you to sign in to Quillm with Google, choose a workspace (or name a new one), and decide whether the agent may change things or only read. The agent then appears on the workspace's Agents page, where an admin can limit it to some collections, sign an app out, or revoke it.

Without a browser (CI, cron, servers)

An admin can create a token for an agent on the workspace's Agents page and give it to the job directly, for example:

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

Keep the token in your CI's secret store, not in a repository.

What the agent can do

Read the workspace, read and write datasets, create and update pages, test-render a page, share a page by link when asked, leave notes for the next agent, and report a problem with Quillm. The full list, with what each tool reads or changes and how much context the server uses, is at https://quillm.ai/connect#tools.

Privacy

What Quillm stores and what it sends to other services is listed in the privacy policy. Workspace admins can export everything at any time.

Files

FileFor
.claude-plugin/plugin.json, .claude-plugin/marketplace.json, .mcp.jsonClaude Code, claude.ai and Cowork; also read by Grok Build
.cursor-plugin/plugin.json, mcp.jsonCursor
.codex-plugin/plugin.jsonChatGPT and Codex
.grok-plugin/plugin.jsonGrok Build
plugin.jsonAgent Plugins (portable manifest)
server.jsonThe Official MCP Registry entry, ai.quillm/quillm

Support

Write to [email protected], or open an issue in this repository.

License

MIT. See LICENSE.

Source: README.md at commit a8dcfc3

Tools

0
Tool metadata has not been indexed yet.

Version history

1
  1. v1.0.0LatestOct 3, 2026