Klever

You can use the Klever MCP Server to store, retrieve, and enrich context for Klever blockchain smart contract development. It provides a centralized knowledge base with patterns, best practices, and runtime guidance that you can query from your MCP clients to speed up development, auditing, and deployment workflows.

How to use

You connect to the MCP server from your client to ingest, query, and enhance knowledge relevant to Klever smart contracts. Use the hosted public MCP server for quick access, or run a local MCP server to host and manage your own knowledge base. Your client can perform read operations, bind to knowledge contexts, and request enhancements that improve query results with relevant Klever VM context.

How to install

Prerequisites: you need Node.js and a package manager installed on your system. Ensure you have a working internet connection to fetch dependencies.

# Run the MCP server locally in development mode using npx
npx -y @klever/mcp-server

# Or connect to the hosted public server
claude mcp add -t http klever-vm https://mcp.klever.org/mcp

Configuration and usage notes

The server supports multiple deployment modes and storage backends. You can run in HTTP API mode, local MCP protocol mode (stdio), or as a public hosted MCP server. You can choose in-memory storage for development or Redis storage for production. The server automatically loads the Klever knowledge base, with CPU- and memory-friendly options depending on your storage choice.

Knowledge loading and runtime behavior

For memory storage, knowledge loads automatically at startup and exists only while the server runs. For Redis storage, ingest the knowledge base once, then start the server so the data persists across restarts. You will access smart contract templates, best practices, deployment scripts, error patterns, testing patterns, and API references through queries.

Security and access patterns

Public mode exposes a read-only subset of tools to keep the knowledge base safe while enabling broad access. Write operations and project initialization tools remain disabled in public mode. When hosting privately, you can enable full read/write capabilities as needed.

Context enhancement and integration

You can automatically enrich your queries with relevant Klever VM context using a dedicated enhancement tool. This helps your client workflows retrieve more precise and valuable results by leveraging the comprehensive knowledge base.

Integration examples

Integrate with editors and tooling to fetch context and suggestions. For example you can query for token transfer examples or run helper scripts that interact with deployed contracts, guided by enriched context from the knowledge base.

Development and testing workflows

Development flows include loading the knowledge base in memory for quick iteration or using Redis for production-grade persistence. You can build, test, ingest, and run in development mode, and you have access to deployment, upgrade, and query scripts for contract management.

API endpoints and examples

The server exposes endpoints for creating, querying, and managing context. You can ingest new contexts, query by filters, retrieve specific contexts, and find contexts similar to a given entry. Batch ingestion, context updates, and deletion are supported in non-public modes.

Ingesting contracts and knowledge base

Use built-in ingestion utilities to parse Klever contracts, extract patterns, and store them in the chosen storage backend. You can ingest individual contracts, entire directories, and common patterns to populate the knowledge base.

Contract validation and quality checks

The server can automatically validate Klever contracts and detect issues such as event annotation formats, API parameter handling, and storage mapper choices. These validations help improve contract quality and reduce common errors.

Project initialization and helper scripts

If you start a new Klever contract project, you can generate helper scripts that support building, deploying, upgrading, querying, testing, and interacting with contracts. This streamlines project setup and ongoing management.

Enhancement and integration workflow

You can enrich queries with context automatically before processing, ensuring you always work with the most relevant Klever VM information. This pattern improves the usefulness of responses across editors and tooling.

Troubleshooting tips

If you encounter startup or connectivity issues, verify storage configuration, ensure the knowledge base ingestions completed successfully, and confirm that the MCP endpoint is reachable from your client. For public mode, confirm you are using the correct public server URL and that write operations are disabled.

Notes

In production, consider placing the server behind a reverse proxy for TLS termination and apply appropriate access controls. Public mode emphasizes read-only operations to minimize risk.

Context types

The system categorizes contexts into types such as code examples, best practices, security tips, optimizations, documentation, error patterns, deployment tools, and runtime behavior explanations.

Pre-loaded knowledge base

The MCP server ships with a comprehensive knowledge base containing entries across multiple categories, including patterns, contract examples, development tools, storage and optimization guidance, best practices, and security considerations.

Usage patterns for developers

Developers can use the MCP server to accelerate development with context-aware suggestions, automate code reviews against best practices and security patterns, and support learning and documentation generation from contract examples.

Available tools

query_context

Search for relevant Klever development context in the knowledge base

add_context

Ingest new context into the knowledge base

get_context

Retrieve a specific context by ID

find_similar

Find contexts similar to a given context

get_knowledge_stats

Get statistics about the knowledge base

init_klever_project

Initialize a Klever smart contract project with helper scripts

enhance_with_context

Automatically enhance queries with relevant Klever VM context