MariaDB MCP connects Claude to a live database and exposes schema inspection plus SQL execution tools., The default read-only mode allows SELECT, SHOW, DESCRIBE, DESC, and USE-style queries, but the README warns that database privileges are the only reliable way to guarantee read-only access., The server includes a `create_database` tool and optional vector-store tools that can create, insert into, search, and delete vector-store tables when enabled., If `MCP_READ_ONLY=false` or credentials have broad privileges, model-generated SQL can create, modify, delete, or expose database state., The server checks for risky FILE privilege behavior, but teams should revoke FILE and other unnecessary global privileges from the connected MariaDB user., HTTP and SSE transports require explicit authentication and restricted host/origin settings before any non-local use., Embedding providers can receive document text or derived content when vector-store tooling is enabled.
Privacy notes
Database credentials, hostnames, database names, schemas, table names, column names, SQL text, query results, errors, and log files may be visible to the MCP client and model provider., MariaDB data can include customer records, credentials, audit logs, business metrics, payment data, healthcare data, or other regulated information., Vector-store tables can persist source documents, embeddings, metadata, and semantic-search results inside MariaDB., OPENAI_API_KEY, GEMINI_API_KEY, HF_MODEL settings, SSL certificate paths, dotenv files, and database passwords should stay out of prompts, issues, logs, screenshots, and committed files., Review log retention because the server writes logs to `logs/mcp_server.log` by default.
Author
MariaDB
Submitted by
oktofeesh1
Claim status
unclaimed
Last verified
2026-06-06
Decision playbook
Review trust signals before you adopt
Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.
Compare context
Selected
0
Current score
63
Baseline
—
Delta
No baseline selected
No major trust-signal divergence detected in the current selection.
Source and provenance checks
Needs review
Confirm ownership and provenance before trusting install instructions.
Source link availableRequired
Open the canonical repository and verify ownership.
Done
Source provenance statusRequired
Marked as source-backed.
Done
Metadata reviewed
No reviewed flag detected in metadata.
Pending
Safety and privacy checks
Complete
Validate risk disclosures before installation or API wiring.
Safety notes presentRequired
Review the listed safety guidance before running commands.
Done
Privacy notes presentRequired
Review data handling notes before connecting accounts or secrets.
Done
Trust level risk gateRequired
Trust level does not block evaluation.
Done
Package and install checks
Needs review
Check package metadata and artifact integrity signals.
Install payload available
Install or copy payload is available for review.
Done
Package verification flag
No package verification flag provided.
Pending
Checksum metadata
No checksum provided for downloaded artifact.
Pending
Compare-driven decision checks
Needs review
Use compare context to validate trade-offs before adoption.
Compare tray has multiple entries
Add at least one more entry to compare trust differences.
7 safety and 5 privacy notes across 6 risk areas. Review closely: credentials & tokens, network access, third-party handling.
6 areas
SafetyGeneralMariaDB MCP connects Claude to a live database and exposes schema inspection plus SQL execution tools.
SafetyGeneralThe default read-only mode allows SELECT, SHOW, DESCRIBE, DESC, and USE-style queries, but the README warns that database privileges are the only reliable way to guarantee read-only access.
SafetyData retentionThe server includes a `create_database` tool and optional vector-store tools that can create, insert into, search, and delete vector-store tables when enabled.
SafetyCredentials & tokensIf `MCP_READ_ONLY=false` or credentials have broad privileges, model-generated SQL can create, modify, delete, or expose database state.
SafetyLocal filesThe server checks for risky FILE privilege behavior, but teams should revoke FILE and other unnecessary global privileges from the connected MariaDB user.
SafetyNetwork accessHTTP and SSE transports require explicit authentication and restricted host/origin settings before any non-local use.
SafetyThird-party handlingEmbedding providers can receive document text or derived content when vector-store tooling is enabled.
PrivacyCredentials & tokensDatabase credentials, hostnames, database names, schemas, table names, column names, SQL text, query results, errors, and log files may be visible to the MCP client and model provider.
PrivacyCredentials & tokensMariaDB data can include customer records, credentials, audit logs, business metrics, payment data, healthcare data, or other regulated information.
PrivacyData retentionVector-store tables can persist source documents, embeddings, metadata, and semantic-search results inside MariaDB.
PrivacyCredentials & tokensOPENAI_API_KEY, GEMINI_API_KEY, HF_MODEL settings, SSL certificate paths, dotenv files, and database passwords should stay out of prompts, issues, logs, screenshots, and committed files.
PrivacyData retentionReview log retention because the server writes logs to `logs/mcp_server.log` by default.
Safety notes
MariaDB MCP connects Claude to a live database and exposes schema inspection plus SQL execution tools.
The default read-only mode allows SELECT, SHOW, DESCRIBE, DESC, and USE-style queries, but the README warns that database privileges are the only reliable way to guarantee read-only access.
The server includes a `create_database` tool and optional vector-store tools that can create, insert into, search, and delete vector-store tables when enabled.
If `MCP_READ_ONLY=false` or credentials have broad privileges, model-generated SQL can create, modify, delete, or expose database state.
The server checks for risky FILE privilege behavior, but teams should revoke FILE and other unnecessary global privileges from the connected MariaDB user.
HTTP and SSE transports require explicit authentication and restricted host/origin settings before any non-local use.
Embedding providers can receive document text or derived content when vector-store tooling is enabled.
Privacy notes
Database credentials, hostnames, database names, schemas, table names, column names, SQL text, query results, errors, and log files may be visible to the MCP client and model provider.
MariaDB data can include customer records, credentials, audit logs, business metrics, payment data, healthcare data, or other regulated information.
Vector-store tables can persist source documents, embeddings, metadata, and semantic-search results inside MariaDB.
OPENAI_API_KEY, GEMINI_API_KEY, HF_MODEL settings, SSL certificate paths, dotenv files, and database passwords should stay out of prompts, issues, logs, screenshots, and committed files.
Review log retention because the server writes logs to `logs/mcp_server.log` by default.
Prerequisites
Python 3.11 and uv available to the MCP client runtime.
Reviewed checkout of the MariaDB MCP repository with dependencies installed.
MariaDB database host, port, database name, username, and password.
Least-privilege MariaDB user scoped to only the databases and operations Claude should access.
Keep `MCP_READ_ONLY=true` for exploration workflows unless a human explicitly approves write-capable tools.
Optional OpenAI, Gemini, or HuggingFace embedding configuration only when vector-store tools are needed.
MariaDB MCP Server is the official MariaDB Model Context Protocol server for
connecting Claude and other MCP clients to MariaDB databases. It exposes tools
for listing databases and tables, inspecting schemas, running read-only SQL, and
optionally managing embedding-backed vector stores inside MariaDB.
The server is implemented with FastMCP and asyncmy. It runs from a Python
checkout with uv, supports stdio by default, and can also run over SSE or
streamable HTTP when operators configure authentication and network controls.
These sources were reviewed on 2026-06-06. Prefer the live repository,
README, package metadata, server implementation, configuration module, and
license for current installation steps, environment variables, tool behavior,
transport options, and safety notes.
Features
Official MariaDB MCP repository under MariaDB/mcp.
Python 3.11 server managed with uv.
Stdio transport by default, with optional SSE and streamable HTTP modes.
Database listing and table listing tools.
Schema inspection with column metadata and foreign-key relationships.
execute_sql tool for read-only SQL workflows when MCP_READ_ONLY=true.
create_database tool for database creation workflows.
Optional vector-store tools when an embedding provider is configured.
OpenAI, Gemini, and HuggingFace embedding provider support.
SSL/TLS environment variables for MariaDB connections.
Host, origin, log, pool-size, and read-only environment configuration.
Installation
Clone the MariaDB MCP repository, install dependencies with uv, create a
reviewed .env, and configure the MCP client to launch the server from that
checkout:
Keep MCP_READ_ONLY=true for exploration, and enforce least privilege at the
MariaDB user level before connecting important data.
Use Cases
Ask Claude to list accessible databases before choosing a target.
Inspect table schemas and foreign-key relationships before writing SQL.
Run reviewed SELECT queries against development or analytics databases.
Compare schema shape across MariaDB environments.
Prototype vector search over approved MariaDB tables.
Use OpenAI, Gemini, or HuggingFace embeddings for semantic search when that data flow is approved.
Keep database creation and vector-store mutation workflows behind human review.
Safety and Privacy
MariaDB MCP is a database control surface. Treat every query as executable code,
use dedicated least-privilege credentials, and rely on database permissions
rather than client-side filters alone. Keep read-only mode enabled by default,
review generated SQL before execution, revoke global FILE privileges, and avoid
connecting production or regulated datasets unless the MCP client, model
session, and operators are approved for that access.
Vector-store features can persist documents, embeddings, and metadata in
MariaDB, while external embedding providers may receive document text or derived
content. Review provider terms, retention, logging, and data residency before
turning those tools on.
Disclosure
MariaDB offers open-source and commercial database products and services. This
listing is not sponsored, paid, or affiliate-driven, and it is scoped to the
source-backed open repository for the MariaDB MCP server.
Duplicate Check
Existing content includes DBHub, a multi-database MCP server that can connect to
MariaDB among other engines. This entry is distinct because it covers
MariaDB/mcp, the official MariaDB-specific MCP server. No dedicated MariaDB
MCP Server, MariaDB/mcp, or matching source URL entry was found in
content/mcp.
Show that MariaDB MCP Server is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.
[](https://heyclau.de/entry/mcp/mariadb-mcp-server)
How it compares
MariaDB MCP Server side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.
1 trust signal differ across this comparison (Submitter).
MCP server from Zilliz for connecting Claude to Milvus vector database collections, text search, vector search, hybrid search, inserts, deletes, indexes, collection loading, database switching, and collection metadata.
Official Pinecone Developer MCP server that connects Claude and other MCP clients to Pinecone projects and documentation for index management, record upserts, semantic search, cascading multi-index search, reranking, and documentation lookup over integrated-inference indexes.
Connect Claude to a Weaviate vector database — run hybrid search, inspect collection config, list tenants, and upsert objects — using Weaviate's built-in Model Context Protocol server.
✓MariaDB MCP connects Claude to a live database and exposes schema inspection plus SQL execution tools.
The default read-only mode allows SELECT, SHOW, DESCRIBE, DESC, and USE-style queries, but the README warns that database privileges are the only reliable way to guarantee read-only access.
The server includes a `create_database` tool and optional vector-store tools that can create, insert into, search, and delete vector-store tables when enabled.
If `MCP_READ_ONLY=false` or credentials have broad privileges, model-generated SQL can create, modify, delete, or expose database state.
The server checks for risky FILE privilege behavior, but teams should revoke FILE and other unnecessary global privileges from the connected MariaDB user.
HTTP and SSE transports require explicit authentication and restricted host/origin settings before any non-local use.
Embedding providers can receive document text or derived content when vector-store tooling is enabled.
✓Milvus MCP can read collection metadata, query collections, and run text, vector, text-similarity, multi-vector, and hybrid searches.
Write-capable tools can create collections, insert data, upsert data, delete entities, create indexes, bulk insert records, load collections, release collections, and switch databases.
The README notes that the environment file has higher priority than command-line arguments, so stale or unexpected environment settings can silently change the target Milvus instance.
SSE and Streamable HTTP modes can expose database operations over HTTP and should be network-restricted.
Remote Milvus or Zilliz Cloud credentials should be scoped to the collections and databases Claude is allowed to access.
✓The server can create indexes and upsert records, so an agent with a write-capable API key can change live Pinecone project state.
Run the npm package `@pinecone-database/mcp` through `npx`, which downloads and executes the published package on each launch; pin to a trusted version if reproducibility matters.
Scope the Pinecone API key to the intended project and use read-only or least-privilege keys when index creation and writes are not needed.
Require human review before `create-index-for-model` and `upsert-records` runs that mutate production indexes.
✓The MCP server runs inside your Weaviate instance and respects its existing RBAC; scope the API key to least privilege.
The object-upsert tool writes data — restrict write access to the collections Claude should modify.
Privacy notes
✓Database credentials, hostnames, database names, schemas, table names, column names, SQL text, query results, errors, and log files may be visible to the MCP client and model provider.
MariaDB data can include customer records, credentials, audit logs, business metrics, payment data, healthcare data, or other regulated information.
Vector-store tables can persist source documents, embeddings, metadata, and semantic-search results inside MariaDB.
OPENAI_API_KEY, GEMINI_API_KEY, HF_MODEL settings, SSL certificate paths, dotenv files, and database passwords should stay out of prompts, issues, logs, screenshots, and committed files.
Review log retention because the server writes logs to `logs/mcp_server.log` by default.
✓Milvus collections can contain embeddings, sparse vectors, scalar fields, IDs, document chunks, metadata, image or multimodal references, query logs, and retrieval results that reveal sensitive project or user data.
Milvus URI, tokens, database names, collection names, vector payloads, filter expressions, and retrieved records should stay out of prompts, issues, logs, screenshots, and committed files.
Search results can include private source content that may be re-exposed in model transcripts or downstream tickets.
HTTP transports, debug tools, query traces, failed-search artifacts, backups, and benchmark datasets need retention and access-control review.
✓The `PINECONE_API_KEY` is read from the MCP client environment and grants access to the associated Pinecone project; keep it out of prompts, notes, and committed files.
Index names, configurations, namespaces, statistics, record contents, and search queries can be exposed to the MCP client and model provider.
Records and search text may contain embedded documents, customer data, or proprietary content, so review what is sent into indexes and returned by searches.
Documentation search and tool calls reach Pinecone endpoints such as api.pinecone.io, so network access and request metadata leave the local machine.
✓Query text, retrieved objects, and collection metadata enter the MCP client context and the model's prompt.
The Weaviate endpoint URL and API key are secrets — keep them in the client config or environment, not in shared repositories.
Prerequisites
Python 3.11 and uv available to the MCP client runtime.
Reviewed checkout of the MariaDB MCP repository with dependencies installed.
MariaDB database host, port, database name, username, and password.
Least-privilege MariaDB user scoped to only the databases and operations Claude should access.
Python 3.10 or newer.
uv installed for the README's recommended run path.
Running local or remote Milvus instance.
Milvus URI, token, and database name selected for the target environment.
Node.js 18 or newer with `npx` available.
Pinecone account and API key generated from the Pinecone console at app.pinecone.io.
MCP client that supports stdio server configuration, such as Claude Desktop or Cursor.
Awareness that index management and record tools require integrated-inference indexes.
A Weaviate instance on v1.37.1 or later (self-hosted or Weaviate Cloud).
The MCP server enabled on that instance via MCP_SERVER_ENABLED=true.
A Weaviate API key with the RBAC permissions for the collections Claude should reach.
An MCP client such as Claude Code or Claude Desktop.
Install
uv run server.py
git clone https://github.com/zilliztech/mcp-server-milvus.git && cd mcp-server-milvus && uv run src/mcp_server_milvus/server.py --milvus-uri <milvus-uri>
npx -y @pinecone-database/mcp
claude mcp add --transport http weaviate https://<your-weaviate-host>/v1/mcp