Exposes Appwrite's API including databases, users, functions, teams, and storage as tools for AI clients.
Also known as appwrite/mcp-for-api, mcp-server-appwrite
- Stars
- 72
- Tools
- —
- Downloads, 30 days
- 334,841
- Latest version
- 0.2.8
- Last commit
- Nov 13, 2026
- License
- MIT
What is Appwrite MCP?
Appwrite exposes Appwrite's API including databases, users, functions, teams, and storage as tools for AI clients.
Summary generated from the project's README
About Appwrite MCP server
Managing Appwrite backend services through an AI client is what this server enables for developers. It exposes core Appwrite API functionalities, including databases, users, functions, teams, and storage, as actionable tools. By connecting this server, an AI agent can interact directly with your Appwrite project to perform administrative or developmental tasks.
Users can connect to the hosted Appwrite Cloud server via a remote URL and authenticate using a browser-based OAuth flow, which removes the need to manually copy API keys. For those using self-hosted Appwrite instances, the server can be run locally over stdio using a project API key. It is compatible with a wide range of clients such as Claude Desktop, Cursor, VS Code, Windsurf, and Zed. Every change made by an agent is recorded with a reason in the Appwrite audit log for easy review.
What you can do with it
- 1Manage backend resourcesAn AI agent can interact with your databases, users, functions, teams, and storage directly.
- 2Audit agent activityYou can track why an AI made specific changes by reviewing the project activity log.
- 3Operate across environmentsThe agent can perform actions within specific project environments like staging.
- 4Search technical documentationDevelopers can use the agent to find information within the Appwrite documentation.
Written from the project's README. Every use case is checked against a quote from it.
How to install Appwrite MCP
Formats verified 10 Nov 2026Install Appwrite MCP in Claude Code
$ claude mcp add mcp-for-api -- uvx mcp-server-appwriteInstall Appwrite MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"mcp-for-api": {
"command": "uvx",
"args": [
"mcp-server-appwrite"
]
}
}
}Install Appwrite MCP in VS Code
.vscode/mcp.json
{
"servers": {
"mcp-for-api": {
"type": "stdio",
"command": "uvx",
"args": [
"mcp-server-appwrite"
]
}
}
}Install Appwrite MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"mcp-for-api": {
"command": "uvx",
"args": [
"mcp-server-appwrite"
]
}
}
}Install Appwrite MCP in Codex
~/.codex/config.toml
[mcp_servers.mcp-for-api] command = "uvx" args = ["mcp-server-appwrite"]
Install Appwrite MCP in Gemini CLI
~/.gemini/settings.json
{
"mcpServers": {
"mcp-for-api": {
"command": "uvx",
"args": [
"mcp-server-appwrite"
]
}
}
}Automated checks
Ran Nov 12, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work
- Source repository linked and reachablePass
- Open source license (OSI)Pass
- Commit or release in the last yearPass
- Package name not close to a popular packagePass
- Remote endpoint uses HTTPSNot checked
- Destructive tools are labelledNot checked
- No hidden instructions to the modelPass
- No critical advisory in the package or its dependenciesPass
README
View on GitHubAppwrite MCP server
A Model Context Protocol server for Appwrite. It exposes Appwrite's API — databases, users, functions, teams, storage, and more — as tools your MCP client can call.
Connect to the hosted server at https://mcp.appwrite.io/ and authenticate
through your browser. The first time you connect, your client opens an Appwrite
consent screen; approve the scopes and you're connected. There are no keys to
copy. The conventional https://mcp.appwrite.io/mcp URL is also supported and
connects to the same server.

Connect your client
Pick your client below. Each adds the hosted Appwrite Cloud server.
claude mcp add --transport http appwrite https://mcp.appwrite.io/
Then, inside a Claude Code session, run /mcp, select appwrite, and follow
the browser prompt to authenticate.
Go to Settings → Connectors → Add custom connector and paste
https://mcp.appwrite.io/. Available on Pro and Max plans; on Team and
Enterprise plans only an organization Owner can add custom connectors.
If you don't see that option (free plan, or a Team/Enterprise member), bridge the remote server through stdio instead (requires Node.js). Go to Settings → Developer → Local MCP servers, click Edit Config, and add:
{
"mcpServers": {
"appwrite": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.appwrite.io/"]
}
}
}
Restart Claude Desktop; the server appears under Local MCP servers and a browser window opens to authenticate.

Edit ~/.cursor/mcp.json (global) or .cursor/mcp.json (project).
{
"mcpServers": {
"appwrite": {
"url": "https://mcp.appwrite.io/"
}
}
}
Cursor prompts you to log in through the browser; the server then shows up under Settings → MCP with its tools enabled.

Edit .vscode/mcp.json (workspace) or your user configuration via the Command
Palette → MCP: Open User Configuration.
{
"servers": {
"appwrite": {
"type": "http",
"url": "https://mcp.appwrite.io/"
}
}
}
Edit ~/.codex/config.toml.
[mcp_servers.appwrite]
url = "https://mcp.appwrite.io/"
Then authenticate from the terminal:
codex mcp login appwrite
In the Codex GUI, you can instead add the server from the MCP settings —
set the URL to https://mcp.appwrite.io/ and leave the token and header
fields empty (authentication happens through the browser):

Edit opencode.json (project) or ~/.config/opencode/opencode.json (global).
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"appwrite": {
"type": "remote",
"url": "https://mcp.appwrite.io/",
"enabled": true
}
}
}
Edit ~/.codeium/windsurf/mcp_config.json.
{
"mcpServers": {
"appwrite": {
"serverUrl": "https://mcp.appwrite.io/"
}
}
}
gemini mcp add --transport http appwrite https://mcp.appwrite.io/
Or edit ~/.gemini/settings.json (note the key is httpUrl, not url):
{
"mcpServers": {
"appwrite": {
"httpUrl": "https://mcp.appwrite.io/"
}
}
}
Gemini CLI opens the browser OAuth flow automatically on first connect. To
re-authenticate, run /mcp auth appwrite inside a session.
Edit ~/.gemini/config/mcp_config.json (global) or .agents/mcp_config.json (project workspace).
{
"mcpServers": {
"appwrite": {
"serverUrl": "https://mcp.appwrite.io/"
}
}
}
⚠️ Note: Antigravity strictly requires the
serverUrlkey for remote transport. Using legacy fields likeurlorhttpUrlwill cause tool registration to fail silently.
Antigravity opens the browser OAuth flow automatically on first connect. If you manually edit the JSON file, navigate to Settings → Customizations → Installed MCP Servers and click Refresh to reload the tool definitions.
copilot mcp add --transport http appwrite https://mcp.appwrite.io/
Or run /mcp add inside a session, or edit ~/.copilot/mcp-config.json:
{
"mcpServers": {
"appwrite": {
"type": "http",
"url": "https://mcp.appwrite.io/"
}
}
}
A browser window opens to authenticate on first connect. Check status with
/mcp.
Go to Settings → AI → MCP Servers → Add Server → Add Remote Server, or add
to your settings.json (zed: open settings):
{
"context_servers": {
"appwrite": {
"url": "https://mcp.appwrite.io/"
}
}
}
Zed prompts you to authenticate through the browser on first connect.
Go to Settings → Agents → MCP servers → + Add, choose the URL-based
server type, and enter https://mcp.appwrite.io/.
Warp opens a browser window to authenticate on first connect.
JetBrains IDEs don't yet support OAuth for remote MCP servers, so bridge through stdio (requires Node.js). Go to Settings → Tools → AI Assistant → Model Context Protocol (MCP) → Add, switch to the JSON view, and paste:
{
"mcpServers": {
"appwrite": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.appwrite.io/"]
}
}
}
A browser window opens to authenticate on first connect.
Cline doesn't yet support OAuth for remote MCP servers, so bridge through stdio (requires Node.js). In the Cline panel, open the MCP Servers icon → Configure tab → Configure MCP Servers, and add:
{
"mcpServers": {
"appwrite": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.appwrite.io/"]
}
}
}
A browser window opens to authenticate on first connect.
Reasons and environments
Every write the agent makes carries a short reason (why it made the change).
Appwrite records it on the project's audit log, so you can review what your
agent changed, and why, in the project's Activity tab.
On Appwrite Cloud, the agent can act in a project's environments (for example
staging): appwrite_get_context lists them, and each call takes the
environment's ID as environment_id next to the project's project_id. See
Tool surface.
Self-hosted Appwrite
Running your own Appwrite instance? Run the MCP server locally over stdio and
authenticate with a project API key. See docs/self-hosted.md
for per-client setup.
Documentation
- Tool surface — the tools exposed to the model and the internal Appwrite catalog.
- How Cloud authentication works — the OAuth 2.1 flow.
- Documentation search — the in-process
appwrite_search_docstool and how to rebuild its index. - Events — MCP Events: webhook subscriptions to changes in your Appwrite project (in development, not available yet).
- Self-hosted Appwrite — run the server locally with a project API key.
- Local development — running, testing, and debugging the server locally.
- AGENTS.md — full contributor guide and pre-PR checklist.
License
This MCP server is licensed under the MIT License. See the LICENSE file for details.
Appwrite MCP questions
How do I add Appwrite MCP to Claude Code?
Run claude mcp add mcp-for-api -- uvx mcp-server-appwrite in your terminal. Add --scope user to make it available in every project.
How do I add Appwrite MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
How do I authenticate with the Appwrite MCP server?
You authenticate through your browser via an Appwrite consent screen. There are no keys to copy for the hosted version.
Can I use this with a self-hosted Appwrite instance?
Yes, you can run the MCP server locally over stdio and authenticate using a project API key.
How can I track changes made by the AI agent?
Every write action includes a short reason which is recorded in the project's Appwrite audit log for review in the Activity tab.