MCP Logging Assistant

MCP Logs Assistant

Centralized logging and monitoring for the MCP ecosystem.

Overview

The Logs Assistant provides comprehensive logging and aggregation capabilities for all MCP services with integrated analytics.

Features

Log Collection: Aggregate logs from all services with in-memory indexing
Real-time Tail: Follow logs in real-time
Advanced Search: Query logs by service, level, time, correlation ID, or content
Time Aggregation: Aggregate logs by minute, hour, or day with statistics
Error Analysis: Analyze error patterns and trends across services
Service Health: Monitor service health based on log metrics
Anomaly Detection: Detect unusual patterns or spikes in log activity
Correlation Tracking: Trace requests across services with correlation IDs
Export: Export logs to JSON, CSV, or text formats
Rotation & Cleanup: Automatic log rotation and retention management

Installation

npm install

Usage

As MCP Server (Claude Desktop)

Configure in Claude Desktop settings:

{
  "mcpServers": {
    "logs": {
      "command": "node",
      "args": ["/path/to/mcp-logs-assistant/index.js"]
    }
  }
}

As HTTP Service

npm start

The service will be available at http://localhost:9103

Available Tools

tail_logs

Follow logs in real-time from specified services.

await client.callTool({
  name: 'tail_logs',
  arguments: {
    service: 'mcp-gateway-assistant', // optional: specific service
    lines: 50 // number of recent lines to show
  }
});

search_logs

Search through logs with advanced filtering.

await client.callTool({
  name: 'search_logs',
  arguments: {
    query: 'error',
    service: 'mcp-gateway-assistant', // optional
    level: 'error', // optional: debug, info, warn, error
    correlationId: 'req-123', // optional: trace specific request
    startTime: '2024-01-01T00:00:00Z', // optional
    endTime: '2024-01-02T00:00:00Z', // optional
    limit: 100
  }
});

get_log_stats

Get statistics about logs.

await client.callTool({
  name: 'get_log_stats', 
  arguments: {
    service: 'mcp-gateway-assistant', // optional
    period: '1h' // 1h, 24h, 7d, 30d
  }
});

rotate_logs

Manually trigger log rotation.

await client.callTool({
  name: 'rotate_logs',
  arguments: {
    service: 'mcp-gateway-assistant' // optional: rotate specific service
  }
});

get_errors

Get recent errors across services.

await client.callTool({
  name: 'get_errors',
  arguments: {
    limit: 50,
    include_stack: true
  }
});

clear_logs

Clear old logs based on retention policy.

await client.callTool({
  name: 'clear_logs',
  arguments: {
    older_than_days: 30,
    dry_run: true // preview what would be deleted
  }
});

Aggregation Tools

aggregate_logs

Aggregate logs by time period with statistics.

await client.callTool({
  name: 'aggregate_logs',
  arguments: {
    period: 'hour', // minute, hour, day
    startTime: '2024-01-01T00:00:00Z',
    endTime: '2024-01-02T00:00:00Z',
    service: 'mcp-gateway-assistant' // optional
  }
});

analyze_errors

Analyze error patterns and trends.

await client.callTool({
  name: 'analyze_errors',
  arguments: {
    service: 'mcp-gateway-assistant', // optional
    timeRange: 86400000, // 24 hours in ms
    minOccurrences: 2 // minimum occurrences to report
  }
});

get_log_statistics

Get comprehensive log statistics.

await client.callTool({
  name: 'get_log_statistics',
  arguments: {
    service: 'mcp-gateway-assistant' // optional
  }
});

find_correlated_logs

Find all logs with a specific correlation ID.

await client.callTool({
  name: 'find_correlated_logs',
  arguments: {
    correlationId: 'req-123-abc'
  }
});

export_logs

Export filtered logs to file.

await client.callTool({
  name: 'export_logs',
  arguments: {
    format: 'json', // json, csv, text
    query: 'error',
    service: 'mcp-gateway-assistant',
    level: 'error',
    startTime: '2024-01-01T00:00:00Z',
    endTime: '2024-01-02T00:00:00Z',
    limit: 1000
  }
});

get_service_health

Analyze service health based on logs.

await client.callTool({
  name: 'get_service_health',
  arguments: {
    service: 'mcp-gateway-assistant',
    timeRange: 3600000 // 1 hour in ms
  }
});

detect_anomalies

Detect anomalies in log patterns.

await client.callTool({
  name: 'detect_anomalies',
  arguments: {
    service: 'mcp-gateway-assistant', // optional
    sensitivity: 5 // 1-10, higher = more sensitive
  }
});

cleanup_old_logs

Clean up logs older than retention period.

await client.callTool({
  name: 'cleanup_old_logs',
  arguments: {
    dryRun: true // preview without deleting
  }
});

Configuration

Aggregator Options

const aggregator = new LogAggregator({
  logsDir: '/path/to/logs',
  aggregateDir: '/path/to/aggregated',
  maxMemoryLogs: 10000, // Max logs to keep in memory
  indexInterval: 60000, // Re-index every minute
  retentionDays: 30 // Keep logs for 30 days
});

Environment Variables

LOG_DIR=~/Documents/mcp-assistant/logs
RETENTION_DAYS=30
INDEX_INTERVAL=60000
MAX_MEMORY_LOGS=10000

Log Format

All services should use structured logging:

{
  "timestamp": "2024-01-01T12:00:00Z",
  "level": "info",
  "service": "mcp-gateway-assistant",
  "message": "Request processed",
  "metadata": {
    "requestId": "abc123",
    "duration": 45,
    "statusCode": 200
  }
}

Data Storage

Service logs: ~/Documents/mcp-assistant/logs/[service-name]/
Aggregated data: ~/Documents/mcp-assistant/data/logs-assistant/aggregated/
Exported logs: ~/Documents/mcp-assistant/data/logs-assistant/aggregated/export-*.{json,csv,text}
In-memory indices: Service, Correlation ID, Error Type

Development

# Run tests
npm test

# Development mode
npm run dev

# CLI usage
npm run tail -- --service gateway
npm run search -- --query "error" --level error
npm run stats -- --period 24h

Performance Features

Indexing

Builds indices on startup for fast queries
Updates indices incrementally every minute
Maintains separate indices for services, correlations, and errors
Limits in-memory cache to prevent memory issues

Query Optimization

Uses indexed lookups for common queries
Supports time-range filtering
Implements result pagination
Caches frequently accessed data

Use Cases

Troubleshooting Service Issues

// Find all errors for a service
const errors = await analyze_errors({
  service: 'mcp-gateway-assistant',
  timeRange: 3600000 // last hour
});

// Check service health
const health = await get_service_health({
  service: 'mcp-gateway-assistant'
});

Request Tracing

// Trace a request across services
const trace = await find_correlated_logs({
  correlationId: 'req-123'
});

Performance Monitoring

// Get hourly aggregations
const metrics = await aggregate_logs({
  period: 'hour',
  startTime: '2024-01-01T00:00:00Z'
});

// Detect anomalies
const anomalies = await detect_anomalies({
  sensitivity: 5
});

Troubleshooting

Slow Queries

Reduce time range
Add more specific filters
Increase cache size if memory allows

Missing Logs

Check if logs are in expected format
Verify log directory paths
Ensure services are writing logs correctly

High Memory Usage

Reduce maxMemoryLogs setting
Implement more aggressive retention policies
Use export and cleanup features regularly

This server cannot be installed

security - not tested

license - not found

quality - not tested

How are these scores calculated?

local-only server

The server can only run on the client's local machine because it depends on local resources.

Provides comprehensive logging and monitoring capabilities for MCP services with real-time log tailing, advanced search, error analysis, and anomaly detection. Enables centralized log aggregation, correlation tracking, and health monitoring across all MCP ecosystem services.

Related MCP Servers

CloudWatch Logs MCP Server
serkanh
-
security
F
license
-
quality
An MCP server that allows AI assistants to access AWS CloudWatch logs by listing log groups and reading log entries.
Last updated -
25
MCP Log Reader
klara-research
A
security
A
license
A
quality
A specialized MCP server that helps analyze and debug Model Context Protocol logs by providing Claude with direct access to log files across multiple platforms.
Last updated -
1
13
MIT License
CloudWatch MCP Server
CharlieFng
-
security
F
license
-
quality
A simplified MCP server that provides a streamlined way to interact with AWS CloudWatch resources (log groups, log queries, and alarms) through the MCP protocol.
Last updated -
4
Cloud Logging API Server
ag2-mcp-servers
-
security
F
license
-
quality
An MCP server that enables interaction with Google Cloud Logging API, allowing users to write, read, and manage log entries and configurations through natural language.
Last updated -

View all related MCP servers