Skip to main content

9 production-ready MCP servers for AI assistants — database, files, git, monitoring, tasks, email, Slack, calendar + CLI discovery

Project description

Sponsor

MCP Toolkit

9 production-ready MCP (Model Context Protocol) servers + CLI discovery tool + Streamlit playground.

MCP is the open standard for connecting AI assistants to external tools. This toolkit provides ready-to-use servers covering common development workflows: database operations, file management, knowledge search, system monitoring, git analytics, and task tracking.

CI Python Tests License

What This Solves

  • Slow tool integration — Stand up MCP servers in minutes instead of weeks
  • Fragmented workflows — One toolkit for data, files, git, monitoring, and tasks
  • Weak governance — Consistent server specs and auto-discovery for safer deployments

Demo

Live demo: https://ct-mcp-toolkit.streamlit.app

Architecture

flowchart TB
    subgraph Client["Client Layer"]
        CLI["CLI (Click)"]
        Playground["Streamlit Playground"]
    end

    subgraph Core["Core"]
        Registry["Server Registry<br/><i>shared/registry.py</i>"]
    end

    subgraph Servers["MCP Servers (9)"]
        FS["file-organizer<br/>5 tools"]
        GH["git-insights<br/>5 tools"]
        KB["markdown-kb<br/>5 tools"]
        AN["system-monitor<br/>5 tools"]
        DB["sqlite-explorer<br/>6 tools"]
        TK["task-tracker<br/>6 tools"]
        EM["email-sender<br/>5 tools"]
        SL["slack-notifier<br/>6 tools"]
        CA["calendar-sync<br/>6 tools"]
    end

    subgraph Protocol["Protocol Layer"]
        FastMCP["FastMCP v2<br/><i>stdio / SSE transport</i>"]
    end

    subgraph Execution["Execution Layer"]
        Tools["| Tools Execution<br/><i>49 tools across 9 servers</i> |"]
    end

    CLI -->|"list / info / serve"| Registry
    Playground -->|"interactive UI"| Registry
    Registry -->|"auto-discovery"| Servers
    Servers -->|"register tools"| FastMCP
    FastMCP -->|"request / response"| Tools

    style Client fill:#e3f2fd,stroke:#1565c0
    style Core fill:#fff3e0,stroke:#ef6c00
    style Servers fill:#e8f5e9,stroke:#2e7d32
    style Protocol fill:#f3e5f5,stroke:#7b1fa2
    style Execution fill:#fce4ec,stroke:#c62828

Directory Layout

mcp-toolkit/
├── mcp_toolkit/
│   ├── servers/           # 9 MCP server implementations
│   │   ├── sqlite_explorer.py    # SQLite CRUD + schema introspection
│   │   ├── file_organizer.py     # File search, dedup, metadata
│   │   ├── markdown_kb.py        # TF-IDF search over markdown docs
│   │   ├── system_monitor.py     # CPU/memory/disk/network via psutil
│   │   ├── git_insights.py       # Repo analytics + contributor stats
│   │   ├── task_tracker.py       # Task CRUD with dependency graphs
│   │   ├── email_sender.py       # SMTP email with templates
│   │   ├── slack_notifier.py     # Slack messaging + alerts
│   │   └── calendar_sync.py      # Google Calendar integration
│   ├── shared/registry.py        # Server auto-discovery
│   └── cli.py                    # CLI: list, info, config, serve
├── app.py                         # Streamlit playground
├── demo_data/                     # Sample files for file-organizer
├── demo_docs/                     # Sample markdown for markdown-kb
└── tests/                         # 180+ pytest tests

Server Catalog

Server Tools Description
sqlite-explorer 6 SQLite database CRUD and schema introspection
file-organizer 5 Smart file search, deduplication, metadata extraction
markdown-kb 5 Knowledge base search using TF-IDF over markdown documents
system-monitor 5 Real-time CPU, memory, disk, and network monitoring
git-insights 5 Git repository analytics, blame, contributor statistics
task-tracker 6 Task management with priorities and dependency graphs
email-sender 5 SMTP email sending with templates and attachments
slack-notifier 6 Slack channel messaging and alerts
calendar-sync 6 Google Calendar event management

Total: 9 servers, 49 tools

Quick Start

Install

git clone https://github.com/ChunkyTortoise/mcp-toolkit.git
cd mcp-toolkit
pip install -e .

Discover Servers

mcp-toolkit list              # List all 6 servers
mcp-toolkit info sqlite-explorer  # Show server tools and details

Configure with Claude Desktop / Claude Code

mcp-toolkit config sqlite-explorer

Output:

{
  "mcpServers": {
    "sqlite-explorer": {
      "command": "mcp-toolkit",
      "args": ["serve", "sqlite-explorer"]
    }
  }
}

Launch a Server

mcp-toolkit serve sqlite-explorer

Try the Playground

make demo

Server Details

sqlite-explorer

Database operations without writing SQL.

Tool Description
create_database Create DB with tables from schema dict
query Parameterized SELECT queries (rejects mutations)
insert Insert a single row
update Update rows matching WHERE clause
delete Delete rows matching WHERE clause
get_schema Introspect tables, columns, and types

file-organizer

Smart file operations for AI-driven workflows.

Tool Description
search_files Recursive glob with depth control
find_duplicates SHA-256 content deduplication
get_metadata File size, timestamps, MIME, hash
analyze_directory Size breakdown by extension
bulk_rename Regex rename with dry-run safety

markdown-kb

Search knowledge bases using TF-IDF ranking.

Tool Description
index_documents Build search index from markdown files
search Cosine similarity ranked search
get_document Retrieve full document content
list_documents List all indexed documents
get_stats Corpus statistics and vocabulary size

system-monitor

Infrastructure observability for AI agents.

Tool Description
get_cpu_stats Usage percent, cores, frequency
get_memory_stats Total/used/available memory
get_disk_stats Disk usage for any mount point
get_network_stats Network I/O and connection count
health_check Pass/fail against CPU/memory thresholds

git-insights

Repository analytics without complex git commands.

Tool Description
get_repo_stats Commits, contributors, branches
get_commit_history Recent commits with metadata
get_blame Line-by-line authorship
get_contributor_stats Per-author commit counts
find_large_files Files exceeding size threshold

task-tracker

Task management with dependency resolution.

Tool Description
create_task Create task with priority (0-4)
update_task Update title, description, status, priority
delete_task Delete task and clean up dependencies
list_tasks Filter by status and priority
add_dependency Add task-to-task dependency
get_ready_tasks Find unblocked pending tasks

email-sender

Enterprise email automation via SMTP with template support.

Tool Description
send_email Send email with optional attachments
send_template_email Send using predefined templates (welcome, notification, alert)
validate_email_config Test SMTP connection and credentials
get_email_history Retrieve recent send history
send_bulk_email Rate-limited bulk email sending

Environment: SMTP_HOST, SMTP_PORT, SMTP_USERNAME, SMTP_PASSWORD, SMTP_USE_TLS

slack-notifier

Slack integration for team notifications and alerts.

Tool Description
send_message Send messages to channels with Block Kit support
send_alert Formatted alerts with severity levels (info/warning/error/critical)
list_channels List available channels
get_channel_info Get channel metadata
send_thread_reply Reply in message threads
upload_snippet Upload text/code snippets

Environment: SLACK_BOT_TOKEN

calendar-sync

Google Calendar integration for event management.

Tool Description
list_events List events in date range
create_event Create new calendar events
update_event Modify existing events
delete_event Remove calendar events
get_free_slots Find available meeting times
validate_calendar_config Test Google API connection

Environment: GOOGLE_API_KEY or GOOGLE_ACCESS_TOKEN, GOOGLE_CALENDAR_ID

Architecture Decisions

ADR Title Status
ADR-0001 FastMCP v2 Over Raw Protocol Accepted
ADR-0002 Six-Server Modularity Accepted
ADR-0003 Analytics and Cost Tracking Accepted

Benchmarks

See BENCHMARKS.md for methodology and results. Key numbers:

Operation P50 Throughput
JSON-RPC Serialize/Deserialize 0.35ms 2,645 ops/sec
Tool Dispatch + Validation 0.06ms 11,987 ops/sec
Server Registry Lookup 0.04ms 15,562 ops/sec
Capability Negotiation 0.96ms 1,000 ops/sec

Development

make test      # Run 180+ pytest tests
make lint      # Ruff lint + format check
make format    # Auto-fix lint + formatting
make demo      # Launch Streamlit playground

Tech Stack

  • FastMCP v2 — MCP server framework
  • Click — CLI interface
  • Streamlit — Interactive playground
  • psutil — System monitoring
  • GitPython — Git operations
  • scikit-learn — TF-IDF search (markdown-kb)

Key Metrics

Metric Value
MCP Servers 9
Total Tools 49
Test Suite 180+ pytest tests
Python 3.11+
Protocol FastMCP v2 (stdio / SSE)
CI GitHub Actions
License MIT

Service Mapping

  • Service 24: Knowledge Management
  • Service 6: AI-Powered Personal and Business Automation
  • Service 21: Workflow Optimization

Certification Mapping

  • IBM RAG and Agentic AI Professional Certificate
  • Duke University LLMOps Specialization
  • Google Cloud Generative AI Leader Certificate
  • Vanderbilt Generative AI Strategic Leader Specialization

Changelog

See CHANGELOG.md for release history.

Related Projects

  • EnterpriseHub -- Real estate AI platform with BI dashboards and CRM integration
  • ai-orchestrator -- AgentForge: unified async LLM interface (Claude, Gemini, OpenAI, Perplexity)
  • llm-integration-starter -- Production LLM patterns: completion, streaming, function calling, RAG, hardening
  • Portfolio -- Project showcase and services

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

mcp_server_toolkit-0.1.0.tar.gz (44.9 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

mcp_server_toolkit-0.1.0-py3-none-any.whl (35.9 kB view details)

Uploaded Python 3

File details

Details for the file mcp_server_toolkit-0.1.0.tar.gz.

File metadata

  • Download URL: mcp_server_toolkit-0.1.0.tar.gz
  • Upload date:
  • Size: 44.9 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.14.2

File hashes

Hashes for mcp_server_toolkit-0.1.0.tar.gz
Algorithm Hash digest
SHA256 edf8a6f6c56e35dc73ce716124b3845d9b2da7c07e35c1ebb3bb6189185ec5a1
MD5 9a0c211f5c00986d1d96f39e876f382c
BLAKE2b-256 70507509748fcc152de50726d7104565eb7720e78d878c47af58f4fc0e2b5cb2

See more details on using hashes here.

File details

Details for the file mcp_server_toolkit-0.1.0-py3-none-any.whl.

File metadata

File hashes

Hashes for mcp_server_toolkit-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 1d81fe3fb5f905deb76ab0d6156908cf7e09575cf63f48e7c52eafd14cff80c7
MD5 2c439362b2dc4de32a605d457be2cd62
BLAKE2b-256 66787af6af3d3f76356398f7e74d11f62ade98fbb5a02c306092f5993480d6f7

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page