Seq MCP Server
Search and stream events from a Seq server.
Seq MCP Server
A Model Context Protocol (MCP) server that provides tools for searching and streaming events from Seq.
Installation
As a .NET Global Tool (Recommended)
# Install
dotnet tool install -g SeqMcpServer
# Update to latest version
dotnet tool update -g SeqMcpServer
# Uninstall
dotnet tool uninstall -g SeqMcpServer
Requirements
- .NET 9.0 Runtime or SDK
- Seq server (local or remote)
- Valid Seq API key
Quick Start
Development Environment
# Clone the repository
git clone https://github.com/willibrandon/seq-mcp-server
cd seq-mcp-server
# Setup development environment (fully automated)
# PowerShell (Windows)
./scripts/setup-dev.ps1
# Bash (Linux/Mac)
./scripts/setup-dev.sh
# Build and run the MCP server
dotnet build
dotnet run --project SeqMcpServer
The setup script automatically:
- Starts a Seq container on ports 15341/18081
- Configures authentication and creates an API key
- Sets up environment variables
- Creates a
.envfile for the application
Production Deployment
MCP servers are not run directly - they are launched by MCP clients. For production:
- Build and deploy the executable:
dotnet publish -c Release -r win-x64 -p:PublishSingleFile=true
- Configure your MCP client to use the deployed executable:
{
"mcpServers": {
"seq": {
"command": "/path/to/seq-mcp-server",
"env": {
"SEQ_SERVER_URL": "http://your-seq-server:5341",
"SEQ_API_KEY": "your-production-api-key"
}
}
}
}
MCP Tools
The following tools are available through the MCP protocol:
-
SeqSearch- Search Seq events with filters- Parameters:
filter(required): Seq filter expression (use empty string""for all events)count: Number of events to return (default: 100)workspace(optional): Specific workspace to query
- Returns: List of matching events
- Example filters:
""- all events"error"- events containing "error"@Level = "Error"- error level eventsApplication = "MyApp"- events from specific application
- Parameters:
-
SeqWaitForEvents- Wait for and capture live events from Seq (5-second timeout)- Parameters:
filter(optional): Seq filter expressioncount: Number of events to capture (default: 10, max: 100)workspace(optional): Specific workspace to query
- Returns: Snapshot of events captured during the wait period (may be empty if no events match)
- Parameters:
-
SignalList- List available signals (read-only)- Parameters:
workspace(optional): Specific workspace to query
- Returns: List of signals with their definitions
- Parameters:
Claude Desktop Integration
Option 1: Using .NET Global Tool (Recommended)
After installing the global tool, add to your Claude Desktop configuration:
{
"mcpServers": {
"seq": {
"command": "seq-mcp-server",
"env": {
"SEQ_SERVER_URL": "http://localhost:5341",
"SEQ_API_KEY": "your-api-key-here"
}
}
}
}
Option 2: Pre-built Release
Download the latest release for your platform and add to your MCP settings:
{
"mcpServers": {
"seq": {
"command": "C:\\\\Tools\\\\seq-mcp-server.exe",
"args": [],
"env": {
"SEQ_SERVER_URL": "http://localhost:5341",
"SEQ_API_KEY": "your-api-key-here"
}
}
}
}
Option 3: Build from Source
Build a single-file executable (requires .NET 9 runtime):
# Windows
dotnet publish -c Release -r win-x64 -p:PublishSingleFile=true
# macOS
dotnet publish -c Release -r osx-x64 -p:PublishSingleFile=true
# Linux
dotnet publish -c Release -r linux-x64 -p:PublishSingleFile=true
The executable will be in SeqMcpServer/bin/Release/net9.0/{runtime}/publish/
Configuration
The Seq MCP Server uses environment variables for configuration:
SEQ_SERVER_URL: URL of your Seq serverSEQ_API_KEY: API key for accessing Seq (required)SEQ_API_KEY_<WORKSPACE>: Optional workspace-specific API keys (e.g.,SEQ_API_KEY_PRODUCTION)
Workspace Support
The MCP server supports workspace-specific API keys (future feature):
export SEQ_API_KEY="default-key"
export SEQ_API_KEY_PRODUCTION="production-key"
export SEQ_API_KEY_STAGING="staging-key"
Note: Workspace-specific keys are currently designed but not yet implemented in the MCP tools.
Development
Prerequisites
- .NET 9.0 SDK
- Docker (for running Seq locally)
Running Tests
dotnet test
Development
The scripts folder contains automated setup scripts:
-
setup-dev.ps1/setup-dev.sh: Automatically configures your development environment- Starts Seq container with authentication
- Handles initial password setup
- Creates development API key
- Sets environment variables
- Creates
.envfile for the application
-
teardown-dev.ps1/teardown-dev.sh: Cleans up the development environment- Stops and removes containers
- Clears environment variables
For detailed development setup, see docs/DEVELOPMENT.md.
Architecture
This is a pure MCP server implementation that:
- Runs as a stdio-based service (no web server)
- Communicates via JSON-RPC over standard input/output
- Does not log to console to avoid interfering with MCP communication
- Optionally logs to Seq itself for debugging when configured
Self-Logging
The MCP server can log its own operations to Seq when a valid SEQ_SERVER_URL and SEQ_API_KEY are provided. This helps with debugging and monitoring the MCP server itself.
License
MIT License - see LICENSE file for details.
Related Servers
kintone Sample MCP Server
Integrate with kintone data and applications using the Model Context Protocol.
BigQuery
Server implementation for Google BigQuery integration that enables direct BigQuery database access and querying capabilities
Open Formula 1 MCP Server
MCP Server to retrieve and analyze Formula 1 (F1) races, cars, drivers, lap, pit stops and more.
Snowflake Cortex
An experimental MCP server to access Snowflake Cortex insights from your development environment.
CData Bullhorn CRM
A read-only MCP server by CData that enables LLMs to query live data from Bullhorn CRM. Requires the CData JDBC Driver for Bullhorn CRM.
Directus MCP Server
An MCP server for Directus CMS, enabling AI clients to interact with the Directus API.
Python MSSQL MCP Server
A Python MCP server for Microsoft SQL Server, enabling schema inspection and SQL query execution.
Trino MCP Server
A Go implementation of a Model Context Protocol (MCP) server for Trino, enabling LLM models to query distributed SQL databases through standardized tools.
Metabase MCP Server
Integrates AI assistants with the Metabase business intelligence and analytics platform.
MongoDB
Interact with MongoDB databases and MongoDB Atlas.