Clearly

sh.clearlyPublished by clearly.shOfficial publisherv1.1.0Updated Oct 24, 2026

Find and manage Clearly workspace docs, canvases, sheets, decks, projects, boards, and tickets.

VerifiedStreamable HTTPWeb executableProductivity & WorkflowFiles & StorageKnowledge & Memory

Overview

AI-generated overview

Remote MCP server for the Clearly workspace, letting an assistant find, read, write, edit, delete and search docs, canvases, sheets, decks, projects, boards…

What you can do
  • Browse and locate workspace items such as documents, canvases, sheets, decks, projects, boards and tickets
  • Read the contents of a workspace item
  • Create or overwrite workspace content
  • Make targeted edits to existing items
  • Delete workspace items
  • Search and pattern-match across the workspace with grep- and glob-style tools
When to use it
Useful when your team keeps documents, slides, sheets, projects and tickets in Clearly and you want an assistant to look things up, search across them, or make and update content directly from a chat client.
Requirements
A remote streamable HTTP endpoint; no local package to deploy. Sign-in is OAuth 2.1 with dynamic client registration, so most clients only need the server URL. Sign-in asks which workspace and which agent the connection acts as. Optional path forms pin a specific workspace and agent.
Before you install
The connection acts as a chosen agent inside a workspace and includes write, edit and delete tools, so an assistant can change or remove workspace content. One credential is stored per server entry, so multiple workspaces need separate entries. The README's CLI installer pipes a shell script and does not verify a signature or checksum.

Installation

In SourceWeft

  1. Open Clearly 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.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/sh.clearly%2Fclearly/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "clearly": {
      "type": "http",
      "url": "https://relay.clearly.sh/mcp"
    }
  }
}

README

Clearly

Your workspace, addressable by agents. Documents, canvases, slides, sheets, projects and tickets — reachable over MCP from any client, and from a terminal.

This repository has no source code. It is the public home for issues, release notes and security reports. Clearly is a hosted service; the server runs on our infrastructure and the CLI is distributed as a build. If you are looking for a place to file a bug or ask what changed in a release, you are in the right place.

MCP

Clearly is an MCP server, not a plugin — your agent connects out to it. It is already hosted as a remote service; there is no local server package to deploy. On Glama, use Connect or copy the URL below and complete OAuth sign-in.

https://relay.clearly.sh/mcp

Authentication is OAuth 2.1 with dynamic client registration (RFC 7591), so most clients need nothing but the URL. Sign-in asks which workspace and which agent the connection acts as.

Claude Code

bash
claude mcp add --transport http clearly https://relay.clearly.sh/mcpclaude mcp login clearly

There is also a Claude Code plugin that bundles the connector with a set of skills.

Other clients

Any client that speaks streamable HTTP MCP can connect to the same URL. Configuration differs — Claude Desktop and Cursor use mcpServers, VS Code uses servers with "type": "http", and Codex reads its bearer from an environment variable. The MCP docs carry a snippet per client.

Holding more than one workspace or agent at once

A client stores one credential per server entry, so two workspaces behind a single entry collide. Two extra path forms exist for that reason — each is a distinct credential key:

URLMeans
/mcpthe workspace and agent chosen at sign-in
/mcp/w/<workspaceId>pinned to one workspace
/mcp/w/<workspaceId>/a/<agentId>pinned to one workspace and one agent

The agent segment is addressing, never authority — the server refuses it unless it matches the token's own agent.

Protocol

Streamable HTTP. Modern clients can use the stateless 2026-07-28 protocol; legacy clients can negotiate 2025-06-18, 2025-03-26 or 2024-11-05 through initialize. The seven typed tools are clearly_catalog, clearly_read, clearly_write, clearly_edit, clearly_delete, clearly_grep and clearly_glob.

serverInfo.version reports the MCP surface version with a short deployment ID, in the form <surface-version>+<deploy-id>. The surface version changes when the tool contract changes.

CLI

bash
curl -fsSL https://clearly.sh/install.sh | sh

The installer writes beehaven to ~/.local/bin (override with BEEHAVEN_BIN_DIR).

Read it before you run it — install.sh is a plain shell script. Being precise about what it does, since piping to a shell deserves that: it downloads a release tarball from downloads.clearly.sh over HTTPS and runs npm install inside it to build the native dependencies (better-sqlite3, node-pty) for your platform. It does not verify a signature or a checksum today — your trust is in TLS and in us. Node 20+ is required and the script checks for it first.

bash
beehaven login                       # authenticatebeehaven connect home                # pick a workspacebeehaven ls ~                        # your documents, as filesbeehaven grep -ril pricing ~         # search thembeehaven call <action> '{...}'       # any workspace actionbeehaven status                      # connection + auth state

beehaven --help lists the rest.

Reporting things

When filing an MCP issue, beehaven status and the output of the failing initialize are the two most useful things to include. The server reports its build in serverInfo.version (<surface-version>+<deploy-id>), which identifies the tool contract and deployed build you were talking to.

Licence

MIT for the contents of this repository. The Clearly service itself is governed by its terms.

Source: README.md at commit 4531b36

Tools

0
SourceWeft has not read this server's tools yet.

Version history

2
  1. v1.1.0LatestSep 24, 2026
  2. v0.3.2Sep 16, 2026

Score

85% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts