MCP-Powered · AI-Native · Enterprise-Ready

AI-Powered DevSecOps Framework

Seamlessly connect your AI assistant to EY's DevSecOps toolchain. Discover templates, manage pipelines, query documentation, and analyze repositories — all through natural conversation.

Explore Capabilities ↓
0 AI Tools
0 Capability Domains
0 Automated Workflows
0 Knowledge Resources

What is the DSOF MCP Server?

A bridge between AI assistants and EY's DevSecOps ecosystem, powered by the Model Context Protocol (MCP).

💻
Developer Asks a question in natural language
🤖
AI Assistant Copilot, Claude, or any MCP client
⚙️
DSOF MCP Server Routes requests to the right tool
☁️
Cloud Services GitHub, Azure DevOps, Azure AI

The Model Context Protocol (MCP) is an open standard that allows AI assistants to securely call external tools. The DSOF MCP Server implements this protocol to give your AI assistant direct access to DevSecOps operations — no switching between tools, no manual copy-paste, just seamless conversation-driven automation.

Capability Domains

Four integrated domains covering the full DevSecOps lifecycle. Click a domain to see its tools.

AI Tools at a Glance

9 specialized tools organized by capability domain. Click any card to see details.

🚀
Template Catalog
Browse 22+ pipeline templates organized by technology — Angular, .NET, Java, Python, React and more.
Application Onboarding Click to flip →
Template Catalog
InputsNone required — discovers all templates automatically
OutputsCategory, template name, technology, path, and usage notes for each template
When to useWhen onboarding a new application and you need the right pipeline template for your tech stack
list_pipeline_templates
📚
Knowledge Base Sync
Synchronize the DSOF wiki and rebuild the local AI-powered search index for offline documentation queries.
Knowledge & Documentation Click to flip →
Knowledge Base Sync
InputsBranch name (optional), max files safety cap
OutputsFiles indexed count, chunks created, embedding model, snapshot version ID
When to useBefore your first documentation query, or after the wiki has been updated
update_wiki_rag_index
🔎
Smart Documentation Search
Ask questions in natural language and get AI-ranked answers from the DSOF wiki — no keyword guessing needed.
Knowledge & Documentation Click to flip →
Smart Documentation Search
InputsNatural language query, number of results (optional)
OutputsRanked documentation snippets with source paths and relevance scores
When to useWhen you have questions about DSOF processes, IAC components, pipeline configuration, or troubleshooting
query_local_wiki_index
📋
Pipeline Directory
Discover all available DSOF admin pipelines with their purpose, IDs, and supported aliases.
Framework Operations Click to flip →
Pipeline Directory
InputsNone required — lists all 6 admin pipelines
OutputsCanonical name, definition ID, description, and available aliases for each pipeline
When to useFirst step when you want to install, update, or manage DSOF framework components
list_admin_pipelines
🔧
Setup Assistant
Automatically extract the required parameters for any admin pipeline directly from documentation.
Framework Operations Click to flip →
Setup Assistant
InputsPipeline name (canonical or alias)
OutputsParameter list with required flags, descriptions, defaults, and example values
When to useAfter selecting a pipeline from the directory, before launching it — to know exactly what parameters you need
prepare_installation_parameters
▶️
Pipeline Launcher
Trigger any DSOF admin pipeline in Azure DevOps with validated parameters — straight from your AI assistant.
Framework Operations Click to flip →
Pipeline Launcher
InputsPipeline name, parameters JSON, source branch (optional)
OutputsPipeline run ID and direct Azure DevOps URL for monitoring
When to useWhen you're ready to execute a framework installation, update, or maintenance operation
queue_admin_pipeline
📊
Execution Monitor
Track the real-time status of any pipeline run — queued, running, or completed — with duration and results.
Framework Operations Click to flip →
Execution Monitor
InputsRun ID, Azure DevOps project name
OutputsStatus, result, duration, timestamps, and direct Azure DevOps URL
When to useAfter launching a pipeline, to check if it succeeded or diagnose failures
get_pipeline_run_status
🔍
Repository Scanner
Analyze any GitHub repository's tech stack, configuration, dependencies, and DSOF readiness in seconds.
Repository Intelligence Click to flip →
Repository Scanner
InputsGitHub organization and repository name
OutputsFramework detection, SDK version, app type, config files, validation checks with recommendations
When to useWhen assessing a repository before onboarding it to the DSOF framework
github_repo_check
📄
Infrastructure Analyzer
Extract Azure components from IAD architecture PDFs and generate deployment-ready YAML templates.
Repository Intelligence Click to flip →
Infrastructure Analyzer
InputsPath to IAD PDF file
OutputsDetected Azure components, networking, environments, plus deployment YAML templates ready for DSOF pipelines
When to useWhen you have an infrastructure architecture diagram and need to generate DSOF-compatible IaC deployment files
extract_pdf_text

End-to-End Workflows

Pre-built automation sequences that chain multiple tools together for common DevSecOps tasks.

1
Framework Installation
📋
Discover Pipeline Directory
🔧
Configure Setup Assistant
▶️
Launch Pipeline Launcher
📊
Monitor Execution Monitor
2
Repository Assessment
🔍
Scan Repository Scanner
🚀
Recommend Template Catalog
3
Knowledge Query
📚
Sync Knowledge Base Sync
🔎
Search Smart Doc Search
4
Template Discovery
🚀
Browse Catalog Template Catalog
5
Infrastructure Mapping
📄
Upload PDF Infrastructure Analyzer
🔎
Cross-reference Smart Doc Search
📃
Generate YAML Deployment Templates

How It Works

A four-layer architecture connecting your IDE to cloud services through AI and the DSOF MCP Server.

Your IDE
💻 VS Code
🗨️ Claude Desktop
⌨️ Any MCP Client
AI Layer
🤖 GitHub Copilot
🤖 Claude AI
💬 Natural Language Interface
MCP Server
⚙️ Tool Router
🔐 Schema Validation
💾 Local RAG Index
Cloud
🐙 GitHub API
☁️ Azure DevOps
🧠 Azure OpenAI

Who Uses It?

Designed for three key audiences across the DevSecOps lifecycle.

💻
Application Teams
  • Discover the right pipeline template for their tech stack
  • Assess repository readiness before DSOF onboarding
  • Query documentation for configuration guidance
  • Monitor pipeline execution status
🔧
DevOps Engineers
  • Install and update framework components via pipelines
  • Manage admin pipeline parameters and execution
  • Sync and maintain the knowledge base index
  • Generate IaC deployment templates from architecture diagrams
💼
Technical Managers
  • Gain visibility into available framework capabilities
  • Track pipeline execution results and timelines
  • Understand repository readiness across the portfolio
  • Access documentation without deep technical context
0
Pipeline Templates
0
Admin Pipelines
0
Tech Stacks Supported
💡
Offline Knowledge Search

Get Started in 3 Steps

Up and running in minutes — no complex setup required.

1
Install
Install the DSOF MCP Server globally via npm from the Azure Artifacts registry.
npm install -g @ey-attg/dsof-mcp-server
2
Configure
Set your environment variables: GitHub PAT, Azure DevOps PAT, and Azure OpenAI credentials.
DSOF_GITHUB_PAT, DSOF_ADO_PAT, ...
3
Connect
Add the MCP server to your AI client — VS Code, Claude Desktop, or any stdio-based MCP client.
mcp.servers → dsof-mcp-server