logo
Free, unlimited AI code reviews that run on commit
git-lrc git-lrc GitHub Install Now We'd appreciate a star git-lrc - Free, unlimited AI code reviews that run on commit | Product Hunt git-lrc - Free, unlimited AI code reviews that run on commit | Product Hunt

vector_search_engine_service

Facilitates the management and high-speed querying of high-dimensional vector embeddings, encompassing functionalities for vector index creation, data ingestion, and nearest-neighbor retrieval. It natively supports persistent storage of vector representations alongside associated descriptive payload data.

Author

vector_search_engine_service logo

RyanLisse

No License

Quick Info

GitHub GitHub Stars 6
NPM Weekly Downloads 0
Tools 1
Last Updated 2026-02-19

Tags

embeddingscloudsearchesstorage embeddingsembeddings efficientlyvector embeddings

Vector Search Engine MCP Gateway

Synopsis

An implementation of the Model Context Protocol (MCP) gateway interfacing with the LanceDB vector database backend. This server furnishes capabilities for optimized storage, execution of similarity matching routines, and overall administration of vector representations complemented by attendant metadata.

Operational Artifacts

The server exposes persistent vector indexes as manageable operational units: - table://{identifier}: Represents a persistent vector store capable of housing embeddings and their associated descriptive context. - Configurable vector dimensionality setting. - Provisions for textual descriptive attributes. - High-performance mechanism for proximity-based searches.

Interface Endpoints

Index Administration

  • POST /table
  • Provision a novel vector index structure.
  • Request Payload Structure:

    { "name": "primary_index",
    "dimension": 1024
    }

Embedding Operations

  • POST /table/{index_name}/vector
  • Append novel vector datum into the specified index.
  • Request Payload Structure:

    { "vector": [0.9, 0.8, ...],
    "text": "descriptive payload string" }

  • POST /table/{index_name}/search

  • Execute a nearest-neighbor search utilizing a provided query vector.
  • Request Payload Structure:

    { "vector": [0.5, 0.4, ...],
    "limit": 50
    }

Deployment Procedure

bash

Obtain source code repository

git clone https://github.com/yourusername/lancedb_mcp.git cd lancedb_mcp

Install required libraries using uv dependency manager

uv pip install -e .

Integration with Desktop Client

bash

Augment your configuration file (claude_desktop_config.json)

"mcpServers": { "vectorstore": { "command": "uv", "args": [ "run", "python", "-m", "lancedb_mcp", "--db-path", "/var/lib/lancedb_data" ] } }

Development Guidelines

bash

Install supplementary tools for development tasks

uv pip install -e ".[dev]"

Execute verification suites

pytest

Enforce standardized code styling

black . ruff .

Environmental Configuration Variables

  • LANCEDB_URI: Specifies the persistent location for the LanceDB file system structure (default setting: "./vector_store_root")

Licensing Information

Distribution and use of this software are governed by the terms outlined in the MIT License. Consult the LICENSE file for specifics.

See Also

`