Enables AI coding agents to securely access and query SQL and NoSQL databases using plain language via dockit and sqlkit desktop apps.
Also known as geek-fun/data-studio-agent, data-studio-agent, @geek-fun/data-studio-mcp
- Stars
- 3
- Tools
- 2
- Downloads, 30 days
- 291
- Latest version
- 0.1.12
- Last commit
- Nov 15, 2026
- License
- Apache-2.0
What is Data Studio Agent MCP?
Data Studio Agent enables AI coding agents to securely access and query SQL and NoSQL databases using plain language via dockit and sqlkit desktop apps. It exposes 2 tools, such as data_studio__list_connections, data_studio__get_status.
Summary generated from the project's README
About Data Studio Agent MCP server
Data Studio provides a secure way for AI coding agents to interact with various database systems using natural language. It functions as an MCP server that connects to your local desktop applications, specifically sqlkit for SQL databases and dockit for NoSQL sources. This setup allows developers to use tools like Claude Code, Cursor, or Windsurf to query data without exposing sensitive credentials to the LLM.
The server supports over 70 SQL databases including PostgreSQL, MySQL, and Snowflake, alongside NoSQL options like MongoDB, Elasticsearch, and DynamoDB. Because the bridge operates locally on 127.0.0.1, your connection secrets stay within your desktop apps and never reach the AI provider. You can manage risk through a three-tier permission model that includes Read Only, Data Read/Write, and Full Access modes. Destructive actions like DROP or DELETE require explicit user confirmation through the client interface.
What you can do with it
- 1Query SQL databases with natural languageAsk your AI agent to list tables or explain query results from a PostgreSQL instance.
- 2Search NoSQL indicesRequest the last ten orders from a specific Elasticsearch index using plain English.
- 3Explore MongoDB collectionsInstruct an agent to find specific user records within a MongoDB database.
- 4Manage database permissions safelyUse the agent to grant or revoke privileges while maintaining security boundaries.
Written from the project's README. Every use case is checked against a quote from it.
How to install Data Studio Agent MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall Data Studio Agent MCP in Claude Code
$ claude mcp add data-studio-agent -- npx -y @geek-fun/data-studio-mcpInstall Data Studio Agent MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"data-studio-agent": {
"command": "npx",
"args": [
"-y",
"@geek-fun/data-studio-mcp"
]
}
}
}Install Data Studio Agent MCP in VS Code
.vscode/mcp.json
{
"servers": {
"data-studio-agent": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@geek-fun/data-studio-mcp"
]
}
}
}Install Data Studio Agent MCP in VS Code CLI
$ code --add-mcp '{"name":"data-studio-agent","type":"stdio","command":"npx","args":["-y","@geek-fun/data-studio-mcp"]}'Install Data Studio Agent MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"data-studio-agent": {
"command": "npx",
"args": [
"-y",
"@geek-fun/data-studio-mcp"
]
}
}
}Install Data Studio Agent MCP in Codex
$ codex mcp add data-studio-agent -- npx -y @geek-fun/data-studio-mcpInstall Data Studio Agent MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.data-studio-agent] command = "npx" args = ["-y", "@geek-fun/data-studio-mcp"]
Install Data Studio Agent MCP in Gemini CLI
$ gemini mcp add -s user data-studio-agent npx -- -y @geek-fun/data-studio-mcpInstall Data Studio Agent MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"data-studio-agent": {
"command": "npx",
"args": [
"-y",
"@geek-fun/data-studio-mcp"
]
}
}
}Started by HostedMCP on 16 Nov 2026
We ran this command for @geek-fun/data-studio-mcp 0.1.10 in an isolated container with no network and no keys. It answered and listed 2 tools.
Data Studio Agent MCP tools 2
Read from the server with tools/listdata_studio__list_connectionsRead-onlyList all database connections from both dockit and sqlkit. Each entry has { backend, id, name, type } — use id as connection_id when calling database tools. If no connections appear, call data_studio__get_status.
data_studio__get_statusRead-onlyGet Data Studio MCP server status and backend availability. Call this first if tools seem missing or a backend tool fails.
Automated checks
Ran Nov 13, 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 labelledPass
- No hidden instructions to the modelPass
- No critical advisory in the package or its dependenciesPass
README
View on GitHubData Studio Agent
Let your AI coding agent securely access all your databases, in plain language.
Local-first. Enterprise-grade security. Open source.
📖 Product Page · npm · dockit · sqlkit · Releases
English · 简体中文
This repository contains the Data Studio MCP Server, a Model Context Protocol server that gives AI coding agents direct access to your databases through the dockit and sqlkit desktop apps.
- SQL (via sqlkit): 70+ databases (PostgreSQL, MySQL, SQL Server, Oracle, SQLite, DuckDB, ClickHouse, Snowflake, BigQuery, and more)
- NoSQL (via dockit): Elasticsearch, OpenSearch, MongoDB, DynamoDB
Features
- Any AI coding agent. Claude Code, Cursor, Windsurf, OpenCode, Codex, Cline, Pi, Qoder, GitHub Copilot, or any MCP client.
- Any OS. macOS, Windows, Linux.
- Any LLM model. Bring your own provider. No lock-in.
- One MCP server, one config. Routes to both SqlKit (SQL) and DocKit (NoSQL) bridges over localhost.
- Enterprise-grade security. See below.
Quick start
1. Prerequisites
Install and launch dockit and/or sqlkit, add a database connection, and make sure Settings → MCP Bridge → Auto-start is enabled (it is by default). Install both apps for the full SQL + NoSQL tool set.
2. Install the MCP server
npm install -g @geek-fun/data-studio-mcp
Or run it without installing (npx downloads it on first run):
npx -y @geek-fun/data-studio-mcp
3. Add it to your AI tool
OpenAI Codex, one command:
codex mcp add data-studio -- npx -y @geek-fun/data-studio-mcp
Claude Code, one command:
claude mcp add --transport stdio data-studio -- npx -y @geek-fun/data-studio-mcp
Cursor. Create .cursor/mcp.json (project) or ~/.cursor/mcp.json (global):
{
"mcpServers": {
"data-studio": {
"command": "npx",
"args": ["-y", "@geek-fun/data-studio-mcp"]
}
}
}
Windsurf. Create ~/.codeium/windsurf/mcp_config.json (global only):
{
"mcpServers": {
"data-studio": {
"command": "npx",
"args": ["-y", "@geek-fun/data-studio-mcp"]
}
}
}
OpenCode. Add to opencode.json (project) or ~/.config/opencode/opencode.json (global):
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"data-studio": {
"type": "local",
"command": ["npx", "-y", "@geek-fun/data-studio-mcp"],
"enabled": true
}
}
}
DeepSeek Harness. Save the overlay as data-studio.cordis.yml:
- insert:
- id: data-studio-mcp
name: '@deepseek-ai/dsh-mcp-client'
config:
serverName: data-studio
transport: stdio
command: npx
args: ['-y', '@geek-fun/data-studio-mcp']
Run dsh web --patch "$PWD/data-studio.cordis.yml", or merge the insert patch into $DSH_HOME/cordis.patch.yml to keep it across runs. Tools appear as mcp__data-studio__* (DSH is developer preview — its config surface may change).
WorkBuddy / CodeBuddy CLI. One command:
codebuddy mcp add --scope user data-studio -- npx -y @geek-fun/data-studio-mcp
Or add the same mcpServers JSON shown for Cursor above to the project-level .mcp.json (~/.codebuddy/.mcp.json for user scope).
Any other MCP client. Register a stdio server with command npx and args -y @geek-fun/data-studio-mcp.
4. Tune permissions (optional)
Open Settings → MCP Bridge in dockit/sqlkit to control what the agent can do:
| Permission mode | What the agent can do |
|---|---|
| Read Only (default) | Explore schemas, run SELECT queries. No writes. |
| Data Read/Write | INSERT, UPDATE, index operations. No deletes/drops. |
| Full Access | Everything, including DELETE, DROP, TRUNCATE. |
5. Start asking
Use plain language. The agent queries your databases for you:
- "List all tables in my PostgreSQL database"
- "Show me the last 10 orders from the Elasticsearch index
orders*" - "Find all users older than 30 in MongoDB"
- "Run this query and explain the results"
The agent reads schemas, runs queries, and explores your data, then shows you every query it executed.
Enterprise-grade security
The LLM gets broad access to your data, but it never sees your credentials. The policy model gates every capability by risk level.
- Credentials never leave the apps. The LLM only ever sees an opaque
connection_id. Real credentials are resolved inside dockit/sqlkit and never cross the MCP boundary. Your passwords and keys stay on your machine, in your app. - ID-based resource access. Agents access databases strictly by connection ID. Credentials never appear in prompts or tool arguments, so there is no path for the model to obtain or exfiltrate connection secrets.
- Three-tier permission model. Read Only / Data Read-Write / Full Access modes gate every capability by risk level, with per-connection overrides. You can mark any connection read-only or allowlist specific actions.
- Explicit user confirmation. Destructive operations (DELETE, DROP, TRUNCATE) surface as
Askin the policy. The client prompts the user for explicit confirmation before anything destructive runs. - Action-level statement classification. SQL is parsed and classified by statement kind (Read / Write / Delete / DDL) before execution. Write-only tools reject DELETE statements; delete tools reject DDL.
- Local-only bridge. The bridge binds to
127.0.0.1exclusively. It is unreachable from other machines, with no server to host and no API keys to manage.
Tools
All tools follow the data_studio__{app}__{backend}__{action} convention: app is the desktop app exposing the bridge (sqlkit or dockit), and backend is the data source behind it (sqlkit, es, mongo, dynamo). The two server-level tools (data_studio__list_connections, data_studio__get_status) have no app segment. The User confirmation column shows which operations surface an explicit confirmation prompt in your AI client before they run.
The exact tool set advertised at runtime depends on your apps' permission mode and configured connections — call data_studio__get_status to see what is currently available.
| Tool | Backend | Risk | Requires permission | User confirmation |
|---|---|---|---|---|
data_studio__list_connections |
Server | 🟢 Safe | Read Only | No |
data_studio__get_status |
Server | 🟢 Safe | Read Only | No |
data_studio__dockit__dockit__list_connections |
dockit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_databases |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_schemas |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_tables |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__get_schema |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__describe_table |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__explain_query |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_indexes |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_foreign_keys |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_views |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_procedures |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_functions |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_triggers |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__get_table_info |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__get_foreign_keys |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_sessions |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__get_slow_queries |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__list_connections |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__execute_query |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__get_object_ddl |
sqlkit | 🟢 Safe | Read Only | No |
data_studio__sqlkit__sqlkit__execute_write |
sqlkit | 🟡 Elevated | Data Read-Write | No |
data_studio__sqlkit__sqlkit__kill_session |
sqlkit | 🟡 Elevated | Data Read-Write | No |
data_studio__sqlkit__sqlkit__grant_privilege |
sqlkit | 🟡 Elevated | Data Read-Write | No |
data_studio__sqlkit__sqlkit__revoke_privilege |
sqlkit | 🟡 Elevated | Data Read-Write | No |
data_studio__sqlkit__sqlkit__execute_delete |
sqlkit | 🔴 Destructive | Full Access | Yes |
data_studio__sqlkit__sqlkit__execute_ddl |
sqlkit | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__es__search |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__get_document |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__cat_indices |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__get_mapping |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__cat_aliases |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__get_alias |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__count |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__cluster_health |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__cat_nodes |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__cat_shards |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__list_snapshots |
dockit · Elasticsearch | 🟢 Safe | Read Only | No |
data_studio__dockit__es__index_document |
dockit · Elasticsearch | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__es__update_document |
dockit · Elasticsearch | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__es__create_index |
dockit · Elasticsearch | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__es__put_mapping |
dockit · Elasticsearch | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__es__put_alias |
dockit · Elasticsearch | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__es__update_aliases |
dockit · Elasticsearch | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__es__bulk |
dockit · Elasticsearch | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__es__reindex |
dockit · Elasticsearch | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__es__restore_snapshot |
dockit · Elasticsearch | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__es__delete_document |
dockit · Elasticsearch | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__es__delete_by_query |
dockit · Elasticsearch | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__es__delete_index |
dockit · Elasticsearch | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__es__delete_alias |
dockit · Elasticsearch | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__mongo__list_databases |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__list_collections |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__find |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__collection_stats |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__database_stats |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__server_status |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__repl_set_status |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__shard_status |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__count_documents |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__list_indexes |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__sample_documents |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__distinct |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__get_slow_queries |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__list_users |
dockit · MongoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__mongo__aggregate |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__insert_one |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__update_many |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__create_database |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__create_collection |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__update_document |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__rename_collection |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__clone_collection |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__create_index |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__drop_index |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__insert_many |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__find_one_and_update |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__bulk_write |
dockit · MongoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__mongo__delete_many |
dockit · MongoDB | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__mongo__drop_database |
dockit · MongoDB | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__mongo__drop_collection |
dockit · MongoDB | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__mongo__delete_document |
dockit · MongoDB | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__mongo__truncate_collection |
dockit · MongoDB | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__dynamo__execute_query |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__describe_table |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__list_tables |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__query_table |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__scan_table |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__batch_get_items |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__describe_continuous_backups |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__describe_ttl |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__get_table_metrics |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__list_backups |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__describe_backup |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__describe_limits |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__list_tags |
dockit · DynamoDB | 🟢 Safe | Read Only | No |
data_studio__dockit__dynamo__execute_write |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__create_item |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__batch_write_items |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__update_item |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__transact_write_items |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__create_gsi |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__update_gsi |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__create_table |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__update_table_config |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__update_ttl |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__update_pitr |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__update_streams |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__restore_table |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__create_backup |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__tag_resource |
dockit · DynamoDB | 🟡 Elevated | Data Read-Write | No |
data_studio__dockit__dynamo__execute_delete |
dockit · DynamoDB | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__dynamo__delete_item |
dockit · DynamoDB | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__dynamo__delete_gsi |
dockit · DynamoDB | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__dynamo__delete_table |
dockit · DynamoDB | 🔴 Destructive | Full Access | Yes |
data_studio__dockit__dynamo__truncate_table |
dockit · DynamoDB | 🔴 Destructive | Full Access | Yes |
| 117 tools total. Read-only operations run automatically under Read Only mode. Elevated operations (writes, index/schema changes) require Data Read-Write. Destructive operations (DELETE, DROP, TRUNCATE) require Full Access and always surface an explicit user confirmation prompt. |
How it works
code agent (Claude Code / Cursor / OpenCode ...)
|
| MCP stdio protocol
v
@geek-fun/data-studio-mcp ← npm package (pure TypeScript)
|
| HTTP (localhost)
+----------------+----------------+
v v |
dockit:9120 sqlkit:9121 |
(NoSQL bridge) (SQL bridge) |
| | |
v v |
Elasticsearch PostgreSQL |
MongoDB MySQL |
DynamoDB SQL Server |
OpenSearch SQLite |
The MCP server is a thin routing layer. All database drivers, SSH tunnels, and connection management live in the desktop apps, which expose a local HTTP bridge (127.0.0.1 only). The MCP server auto-discovers running backends via each app's port file.
For developers
This repository also contains the data-studio-agent Rust framework, the shared AI agent loop (provider adapters, streaming, tool calling, context compaction) that powers the built-in assistants in dockit and sqlkit.
See crates/data-studio-agent/README.md for installation, architecture, and integration guides.
License
Apache 2.0. See LICENSE.
Data Studio Agent MCP questions
How do I add Data Studio Agent MCP to Claude Code?
Run claude mcp add data-studio-agent -- npx -y @geek-fun/data-studio-mcp in your terminal. Add --scope user to make it available in every project.
How do I add Data Studio Agent MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
What tools does Data Studio Agent MCP provide?
It exposes 2 tools, including data_studio__list_connections, data_studio__get_status.
Which databases are supported?
It supports over 70 SQL databases including PostgreSQL, MySQL, SQL Server, and Snowflake, as well as NoSQL databases like Elasticsearch, MongoDB, and DynamoDB.
Is my database credential safe?
Yes, credentials never leave your local apps. The LLM only receives an opaque connection ID, and real credentials are resolved inside dockit or sqlkit.
Can the AI delete my data?
Destructive operations like DELETE or DROP require Full Access permission and will trigger an explicit user confirmation prompt in your AI client.
What operating systems are supported?
The server works on macOS, Windows, and Linux.