Clearly — your workspace, addressable by agents. Issues, release notes and security reports for the MCP server and the beehaven CLI.
- ✓Open-source license (MIT)
- ✓Actively maintained (<30d)
- ✓Clear description
- ✓Documented (README)
- !Install pipes a remote script into a shell (curl | sh)
/plugin marketplace add clearly-sh/clearly
/plugin install clearlyResumen de Plugins
# Clearly
**Your workspace, addressable by agents.** Documents, canvases, slides, sheets, projects and
tickets — reachable over [MCP](https://modelcontextprotocol.io) from any client, and from a
terminal.
- **Website** — [clearly.sh](https://clearly.sh)
- **MCP docs** — [clearly.sh/docs/mcp](https://clearly.sh/docs/mcp)
- **MCP Registry** — [`sh.clearly/clearly`](https://registry.modelcontextprotocol.io/v0.1/servers?search=sh.clearly)
> **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.
```
https://relay.clearly.sh/mcp
```
Authentication is OAuth 2.1 with dynamic client registration ([RFC 7591](https://datatracker.ietf.org/doc/html/rfc7591)),
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/mcp
claude mcp login clearly
```
There is also a [Claude Code plugin](https://github.com/clearly-sh/clearly-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](https://clearly.sh/docs/mcp) 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:
| URL | Means |
|---|---|
| `/mcp` | the 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. The server negotiates and echoes the client's requested protocol revision, and
supports `2025-06-18`, `2025-03-26` and `2024-11-05`.
---
## 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`](https://clearly.sh/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 # authenticate
beehaven connect home # pick a workspace
beehaven ls ~ # your documents, as files
beehaven grep -ril pricing ~ # search them
beehaven call <action> '{...}' # any workspace action
beehaven status # connection + auth state
```
`beehaven --help` lists the rest.
---
## Reporting things
- **Bugs and feature requests** — [open an issue](https://github.com/clearly-sh/clearly/issues/new/choose).
- **Security** — please do **not** open a public issue. See [SECURITY.md](SECURITY.md).
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`
(`0.3.1+<commit>`), which tells us exactly what you were talking to.
---
## Licence
[MIT](LICENSE) for the contents of this repository. The Clearly service itself is governed by its
[terms](https://clearly.sh/terms).
Lo que la gente pregunta sobre clearly
¿Qué es clearly-sh/clearly?
+
clearly-sh/clearly es plugins para el ecosistema de Claude AI. Clearly — your workspace, addressable by agents. Issues, release notes and security reports for the MCP server and the beehaven CLI. Tiene 0 estrellas en GitHub y su última actualización registrada es del 2026-09-10.
¿Cómo se instala clearly?
+
Puedes instalar clearly clonando el repositorio (https://github.com/clearly-sh/clearly) o siguiendo las instrucciones del README en GitHub. ClaudeWave también te ofrece bloques de instalación rápida en esta misma página.
¿Es seguro usar clearly-sh/clearly?
+
Nuestro agente de seguridad ha analizado clearly-sh/clearly y le ha asignado un Trust Score de 79/100 (tier: Trusted). Revisa el desglose completo de comprobaciones superadas y flags en esta página.
¿Quién mantiene clearly-sh/clearly?
+
clearly-sh/clearly es mantenido por clearly-sh. La última actividad registrada en GitHub es del 2026-09-10, con 0 issues abiertos.
¿Hay alternativas a clearly?
+
Sí. En ClaudeWave puedes explorar plugins similares en /categories/plugins, ordenados por popularidad o actividad reciente.
Despliega clearly en tu cloud
Lleva este repo a producción en minutos. Cada plataforma genera su propio entorno con variables de entorno editables.
¿Mantienes este repo? Añade un badge a tu README
Pega el badge en tu README de GitHub para mostrar que está auditado por ClaudeWave. Cada badge enlaza de vuelta a esta página y muestra el Trust Score actual.
[](https://claudewave.com/repo/clearly-sh-clearly)<a href="https://claudewave.com/repo/clearly-sh-clearly"><img src="https://claudewave.com/api/badge/clearly-sh-clearly" alt="Featured on ClaudeWave: clearly-sh/clearly" width="320" height="64" /></a>Más Plugins
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary
Academic Research Skills for Claude Code: research → write → review → revise → finalize
Agent skill that removes signs of AI-generated writing from text
A skill to stop your coding agent from burying the answer. ADHD-friendly output.
Create beautiful slides on the web using a coding agent's frontend skills