boltmcp
MCP server from boltmcp/boltmcp
claude mcp add --transport stdio boltmcp-boltmcp docker compose -f docker-compose.yml up -d \ --env BOLT_ENV="production" \ --env BOLT_LOG_LEVEL="info"
How to use
BoltMCP is an enterprise-grade orchestration platform for creating, deploying, and managing custom MCP servers on-premises. It provides a self-hosted, cloud-native stack with built-in security and orchestration features, including OAuth 2 based access, streamable HTTP transport, and the ability to register, compose, and transform tools that interact with multiple upstream APIs. The platform ships with an MCP client playground and an MCP Inspector to help you experiment with server configurations, test tool interactions, and inspect input/output transformations in real time. You can leverage BoltMCP to spin up tailored MCP servers on demand, register your own tools, and orchestrate complex API workflows within a declarative specification, enabling repeatable deployments across environments.
How to install
Prerequisites:
- Docker and Docker Compose installed on your host
- Access to a Kubernetes cluster if deploying via Helm (optional for production)
- kubectl configured for your cluster (if using Helm for Kubernetes)
Installation (Docker Compose quickstart):
- Clone the BoltMCP repository or download the docker-compose files from the provided directory.
- Ensure docker-compose.yml is present in your project root.
- Start BoltMCP with Docker Compose: docker compose -f docker-compose.yml up -d
- Verify the services are running: docker compose -f docker-compose.yml ps
Deployment to production (Helm on Kubernetes):
- Install Helm if not already installed.
- Add BoltMCP Helm repo (if applicable) and update repos: helm repo add boltmcp https://example.com/boltmcp/charts helm repo update
- Deploy the BoltMCP chart with your values file: helm install boltmcp boltmcp/boltmcp --values values.yaml
- Monitor deployment and ensure all pods are Healthy: kubectl get pods -A
Post-install:
- Configure OAuth 2 credentials for secure access.
- Register your MCP tools and define transformations via the declarative spec.
- Access the MCP client playground and MCP Inspector from the provided UI endpoints.
Additional notes
Tips and considerations:
- The BoltMCP stack is designed for on-premises deployments; ensure your network policies allow the necessary API calls.
- When using Docker Compose, keep your docker-compose.yml version and service definitions aligned with your environment and Kubernetes integration if you plan to migrate later.
- Enable OAuth 2 by default for all servers to secure access; document and rotate credentials regularly.
- The MCP client playground and Inspector are powerful for validating tool inputs/outputs and should be used during development before moving to production.
- If you need to customize tools, BoltMCP supports declarative specs for transformation chains and lazy-loading of tools to optimize startup times.
Related MCP Servers
PPTAgent
An Agentic Framework for Reflective PowerPoint Generation
mcp-tool-kit
Agentic abstraction layer for building high precision vertical AI agents written in python for Model Context Protocol.
ContextPods
Model Context Protocol management suite/factory. An MCP that can generate and manage other local MCPs in multiple languages. Uses the official SDKs for code gen.
context-engineering
[WIP] Context engineering: the art and science of shaping context-aware AI systems
knowledgebase
BioContextAI Knowledgebase MCP server for biomedical agentic AI
prospectio-api
MCP/API server that helps you to connect to different lead generation app