Apricot MCP public description
- ✓Actively maintained (<30d)
- ✓Clear description
- ✓Documented (README)
- !Licence file present but not machine-readable
git clone https://github.com/metadevpro/apricot-mcp{
"mcpServers": {
"apricot-mcp": {
"command": "node",
"args": ["/path/to/apricot-mcp/dist/index.js"]
}
}
}MCP Servers overview
<!-- mcp-name: tools.apricot/apricot -->
# Apricot MCP

Public information page for the **Apricot** [Model Context Protocol](https://modelcontextprotocol.io)
(MCP) server — the remote MCP endpoint that lets AI coding agents (Claude Code, OpenCode, Codex,
and other MCP clients) manage [Apricot](https://apricot.tools) projects and SysML2 models directly
from a conversation.
This repository does not contain the server's implementation. It exists to give the MCP server a
public, citable home — a discoverable `repository` reference for registries such as the
[official MCP Registry](https://registry.modelcontextprotocol.io) and the
[GitHub MCP Registry](https://github.com/mcp) — while the implementation itself stays in
Apricot's private monorepo.
- **Live endpoint**: `https://mcp.apricot.tools/`
- **Product**: [apricot.tools](https://apricot.tools) — a SysML2 (Systems Modeling Language v2)
editor with an in-editor AI assistant ("Alicia"), by [Metadev](https://metadev.pro).
- **Protocol version**: [2026-07-28](https://modelcontextprotocol.io/docs/2026-07-28/getting-started/intro)
- **Transport**: streamable HTTP, OAuth 2.1 (Authorization Code + PKCE, with Dynamic Client
Registration) — no manual token to create or paste.
> **Note:** the registry namespace above (`tools.apricot/apricot`) reflects the intended naming,
> matching Structura's convention (reverse-DNS of the product domain). Confirm it against the
> actual `server.json` once Apricot MCP is first published to the MCP Registry.
## What it does
- List, load, create, or delete your Apricot projects.
- Browse and read the files of a project.
> Editing and validating a SysML2 model (set current model, delta edits, validation) is currently
> only available through Alicia, Apricot's in-editor assistant — those tools are scope-gated and
> not yet exposed to general MCP clients over this endpoint.
## Tools
The authoritative list is always what the live server reports over `tools/list`; the table below
is a convenience summary and may lag a version behind.
| Tool | What it does |
| --- | --- |
| `list_projects` | List the caller's Apricot projects. |
| `create_project` | Create a new project (starts with no files). |
| `load_project` | Load a project the caller owns, by id. |
| `update_project` | Update a project's name or description. |
| `delete_project` | Permanently delete a project the caller owns. |
| `list_project_files` | List a project's files (name/path/updatedAt, no content). |
| `get_project_file` | Read one file's full content from a project. |
## Install
Any MCP-capable agent can connect. A few examples:
### Claude Code
```bash
claude mcp add --transport http apricot https://mcp.apricot.tools/
```
Then run `/mcp` inside a Claude Code session, select **apricot**, and choose **Authenticate**.
### OpenCode
```bash
opencode mcp add
```
Pick **remote**, name it `apricot`, and paste `https://mcp.apricot.tools/`. Then:
```bash
opencode mcp auth apricot
```
### Codex
```bash
codex mcp add apricot --url https://mcp.apricot.tools/
codex mcp login apricot
```
Each of these opens a browser window with the normal Apricot login (Google/GitHub) — approve it
there, and the agent reuses that session from then on.
## Support
- Product and account questions: [apricot.tools](https://apricot.tools)
- Issues with the MCP server itself: open an issue on this repository.
## License
See [LICENSE](./LICENSE). This repository's own content (this README, the license file) is
provided for informational and discovery purposes; it does not grant any rights to the Apricot
service, product, or its non-public source code.
---
© [Metadev](https://metadev.pro), 2026.
What people ask about apricot-mcp
What is metadevpro/apricot-mcp?
+
metadevpro/apricot-mcp is mcp servers for the Claude AI ecosystem. Apricot MCP public description It has 0 GitHub stars and its last recorded update is dated 2026-09-14.
How do I install apricot-mcp?
+
You can install apricot-mcp by cloning the repository (https://github.com/metadevpro/apricot-mcp) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is metadevpro/apricot-mcp safe to use?
+
Our security agent has analyzed metadevpro/apricot-mcp and assigned a Trust Score of 75/100 (tier: Trusted). See the full breakdown of passed checks and flags on this page.
Who maintains metadevpro/apricot-mcp?
+
metadevpro/apricot-mcp is maintained by metadevpro. The last recorded GitHub activity is dated 2026-09-14, with 0 open issues.
Are there alternatives to apricot-mcp?
+
Yes. On ClaudeWave you can browse similar mcp servers at /categories/mcp, sorted by popularity or recent activity.
Deploy apricot-mcp to your cloud
Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.
Maintain this repo? Add a badge to your README
Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.
[](https://claudewave.com/repo/metadevpro-apricot-mcp)<a href="https://claudewave.com/repo/metadevpro-apricot-mcp"><img src="https://claudewave.com/api/badge/metadevpro-apricot-mcp" alt="Featured on ClaudeWave: metadevpro/apricot-mcp" width="320" height="64" /></a>More MCP Servers
Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
An open-source AI agent that brings the power of Gemini directly into your terminal.
Real-time global intelligence dashboard. AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface
🕷️ An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl! Don't be shy, join here: https://discord.gg/EMgGbDceNQ
The fastest path to AI-powered full stack observability, even for lean teams.