Skip to main content

seekdb Cursor Extension

In the era of AI-assisted programming, developers increasingly rely on intelligent IDEs to improve development efficiency. However, when using specific technology stacks like the seekdb vector database, AI assistants often lack domain-specific knowledge, making it difficult to provide accurate code suggestions and assistance. To address this issue, OceanBase provides the seekdb Cursor Extension, a plugin designed for the Cursor editor that integrates seekdb's official documentation into the Cursor AI assistant, equipping it with the expertise needed for seekdb database development.

Background​

What is Cursor?​

Cursor is an AI-powered code editor that supports Windows, macOS, and Linux operating systems. Cursor comes with a powerful AI assistant that helps developers:

  • Smart code completion: Automatically completes code based on context.
  • Code generation: Generate code from natural language descriptions.
  • Code interpretation and refactoring: Understand and optimize existing code.
  • Q&A interaction: Directly interact with AI to solve programming problems.

Developers can provide additional knowledge and rules to the AI assistant through the .cursor/rules directory, enabling it to better understand specific technology stacks. For more information about Cursor rules, see Cursor rules documentation.

Introduction to seekdb Cursor Extension​

seekdb Cursor Extension is a Cursor editor extension that allows the AI assistant to access seekdb's official documentation by configuring Cursor Rules. This extension enables the AI assistant to:

  • Understand seekdb database concepts: Vector search, hybrid search, AI functions, etc.
  • Provide accurate code suggestions: Generate code that follows best practices based on the official documentation.
  • Answer seekdb-related questions: Directly access technical support within the editor.
  • Accelerate the development process: Reduce the time spent looking up documentation and focus on business logic.

Core components​

File/DirectoryDescription
.cursor/rules/seekdb.mdcThe AI assistant behavior rules file, defining two-layer access strategies and document reference specifications
.cursor/rules/seekdb-docs/Local document cache directory, serving as a backup data source when GitHub is inaccessible

After installing and enabling the extension, it will copy the seekdb.mdc rule file and the complete seekdb documentation to your project's .cursor/rules directory. The AI assistant will prioritize fetching the latest documentation from GitHub; if the network is unavailable, it will automatically switch to the local documentation, ensuring you always have access to help.

Install the plugin​

Install it from the Cursor extension marketplace. Follow these steps:

  1. Open the Cursor editor.

  2. Click the extension icon at the top of the sidebar, or use the shortcut:

    • Windows/Linux: Ctrl+Shift+X
    • macOS: Cmd+Shift+X

  3. In the search box, enter "seekdb".

  4. Find the "seekdb" extension and click Install.

Key features​

One-click addition of seekdb documentation​

You can easily add seekdb's official documentation to your current project with a simple command:

Steps:

  1. Open the command panel:

    • Windows/Linux: Press Ctrl+Shift+P
    • macOS: Press Cmd+Shift+P

  2. Enter "seekdb Docs" or "Add seekdb Docs"

  3. Select the Add seekdb Docs command

The documentation will be automatically added to:

  • .cursor/rules/seekdb-docs directory (official documentation)
  • .cursor/rules/seekdb.mdc file (AI rules file)

Dual-mode document access​

The extension supports two working modes with automatic switching:

ModeTrigger conditionDocument sourceReference format
Remote mode (primary)GitHub is accessibleFetches the latest documentation from GitHub in real timeGitHub link
Local mode (backup)Network timeout, error, or unreachableReads from the local .cursor/rules/seekdb-docs/ directoryLocal path

Workflow​

  1. The AI assistant first tries to fetch the documentation directory from GitHub using curl.
  2. If it successfully returns markdown content → Use remote mode.
  3. If it times out, encounters an error, or returns empty content → Immediately switch to local mode.
  4. Once switched to local mode, the assistant will no longer attempt remote access in the same conversation.

Version management​

  • Automatically detects the documentation version
  • Only copies the documentation when the version updates

One-click removal of documentation​

If you no longer need seekdb documentation, you can easily remove it:

  1. Open the command panel (Ctrl+Shift+P or Cmd+Shift+P).
  2. Enter "Remove seekdb Docs".
  3. Select the Remove seekdb Docs command.

Use cases​

Scenario 1: Quick learning of seekdb​

When you're just starting with seekdb, you can directly ask questions in Cursor:

User: How do I create a vector index in seekdb? AI Assistant: According to the seekdb documentation, you can use the following SQL to create an HNSW vector index...

Scenario 2: Writing vector search code​

While writing code, the AI assistant can provide accurate code examples based on the documentation:

User: Write code for hybrid search in seekdb. AI Assistant:

import pyseekdb

# Connect to seekdb
client = pyseekdb.Client(...)
# Get the collection
collection = client.get_collection("...")
# Perform hybrid search
results = collection.hybrid_search(
   query={
       "where_document": {"$contains": "..."},
       "where": {"category": {"$eq": "..."}},
       "n_results": 10
   },
   knn={
       "query_texts": ["..."],  # Will be embedded automatically
       "where": {"year": {"$gte": ...}},
       "n_results": 10
   },
   rank={"rrf": {}},  # Reciprocal Rank Fusion
   n_results=5,
   include=["documents", "metadatas", "embeddings"]
)

Scenario 3: Debugging and troubleshooting​

When encountering issues, the AI assistant can provide solutions based on the official documentation:

User: My vector search is returning inaccurate results. What could be the cause? AI Assistant: According to the seekdb documentation, common causes include... (provides specific troubleshooting steps and solutions)

Summary​

The seekdb Cursor Extension serves as a bridge between seekdb and the Cursor AI assistant, allowing developers to access seekdb expertise while writing code. Whether you're just starting with seekdb or developing production-level AI applications, this extension significantly enhances your development efficiency.

Contents

Edit this page