Workday Studio MCP

Workday Studio MCP

A local MCP server that gives Claude direct access to a Workday Studio workspace, enabling reading, writing, planning, and validating integration assemblies without copy-pasting XML.

Category
Visit Server

README

Workday Studio MCP

A local MCP (Model Context Protocol) server that gives Claude direct access to your Workday Studio workspace. Read, write, plan, and validate integration assemblies without copy-pasting XML back and forth.

Everything runs locally on your machine — no network calls, no tenant credentials, no shared state. The server only sees the Studio Workspace folder you point it at.


What you get

26 tools across these categories:

Category What it does
Navigation list_studio_projects, list_project_files, read_integration_file, search_studio_files, get_workspace_structure
File management write_integration_file, copy_file_from_project, rename_file, delete_file, validate_xml_file
Project setup create_studio_project, create_xsl_transform
Assembly editing list_assembly_steps, list_integration_params, add_assembly_step, update_sub_flow, rename_steps, delete_assembly_step, validate_assembly
Planning plan_integration — design elicitation + skeleton generator; get_workflow — where an integration stands and what to do next
Reference get_step_type_reference — confirmed step type docs with production XML examples; lookup_soap_operations — WWS service/operation lookup with WSDL links; get_patterns — the curated cross-integration knowledge base (diagram rules, MVEL/XSLT/RAAS idioms) by section or search
Diagnostics parse_server_log — parse a Workday integration server log from ~/Downloads
Knowledge capture log_learning — append a discovered pattern/gotcha to the shared intake log (learnings.md)

A growing knowledge base lives at docs/studio-integration-patterns.md — hard-won lessons captured from real Studio debugging sessions.

For a single-page digest of the tools, the plan_integration design brief, validator rule codes, the SOAP catalog, and the highest-value patterns, see docs/quick-reference.md.


Quick install

One line:

curl -fsSL https://raw.githubusercontent.com/krishnagutta/Workday-studio-mcp/main/bin/quickstart.sh | bash

This clones the repo to ~/Workday-studio-mcp, installs dependencies, and prints the exact claude mcp add command for your machine.

Or do it manually — see Manual setup below.


Prerequisites

  • Node.js 18+ — check with node --version
  • Workday Studio installed with at least one project in your workspace
  • Claude Desktop or Claude Code (CLI) — both work

Manual setup

1. Clone and install

git clone https://github.com/krishnagutta/Workday-studio-mcp.git
cd Workday-studio-mcp
npm install

2. Configure your workspace path

cp config.json.example config.json

Open config.json and set workspace_path to the folder containing your Studio projects (the same one Eclipse opens):

{
  "workspace_path": "/Users/yourname/Documents/Studio Workspace",
  "max_file_size_kb": 500,
  "backup_on_write": true,
  "excluded_dirs": [".git", ".settings", "bin", "build", "node_modules", ".metadata", ".plugins"],
  "excluded_extensions": [".class", ".jar", ".zip", ".bak"]
}

Or skip config.json entirely and use an env var:

export STUDIO_WORKSPACE_PATH="/Users/yourname/Documents/Studio Workspace"

3. Verify it starts

node src/index.mjs

You should see:

[studio-mcp] Server started. Workspace: /Users/yourname/Documents/Studio Workspace

Press Ctrl+C to stop — Claude will spawn it on demand.


Connect to Claude

Option A — Claude Code (CLI)

claude mcp add studio-mcp node /absolute/path/to/Workday-studio-mcp/src/index.mjs

Confirm:

claude mcp list

Option B — Claude Desktop

Edit the config file:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json

Add inside mcpServers:

{
  "mcpServers": {
    "studio-mcp": {
      "command": "node",
      "args": ["/absolute/path/to/Workday-studio-mcp/src/index.mjs"]
    }
  }
}

Save and restart Claude Desktop.

Tip: Use which node to get your full node path if Claude can't find it.


Use it

List projects

"List my Studio projects"

Read a file

"Read the assembly.xml from INT145"

Plan a new integration

"I need to build a new integration"

Claude asks design questions (data source, destination, trigger, record volume, auth, error handling) before generating anything. Then it writes a skeleton assembly.xml + assembly-diagram.xml you can open in Studio immediately.

It also writes an aidlc-docs/ folder at the project root, so the design rationale outlives the conversation:

<your project>/aidlc-docs/
  ├── plan.md      # design brief, per-sub-flow prop contracts, open gaps, tenant handoff checklist
  └── state.json   # lifecycle state — sub-flow status, decisions log

Commit these with your integration: they are the record of why it is shaped this way. Regenerating the scaffold refreshes both files but preserves the created date and the decisions log. Attribute names are recorded, never credential values.

Pick up where you left off

"What's the status of INT999_Employee_Sync?"

get_workflow reads aidlc-docs/state.json and reports the phase, which sub-flows are built vs still TODO stubs, the last validation result, and the single next recommended action — without reading the XML. For a project that predates aidlc-docs/, call it with retrofit=true to derive state from the existing assembly.

Fill in a sub-flow

"Fill in the GetWorkers sub-flow — here's the RAAS sample: [paste XML]"

update_sub_flow surgically replaces the TODO stub with real steps and validates the result.

Search across integrations

"Find all uses of integrationMapLookup"

Validate

"Validate the assembly for INT145"

Returns errors (broken routes, illegal comments, missing attributes) and warnings (missing XSL files, unresolved sub-flow endpoints).

Look up a step type

"Show me the cc:http-out reference"

Returns confirmed XML examples, schema rules, and gotchas.

Rename a step safely

"Rename AsyncMediation3 to SetTransactionProps"

rename_steps updates the step ID in assembly.xml and every href in assembly-diagram.xml atomically — renaming in only one file crashes Studio's diagram view.

Find a SOAP operation

"Which WWS service has Put_Applicant?"

lookup_soap_operations searches the Workday Web Services catalog and returns the service, common operations, and WSDL link.

Pull curated guidance before hand-editing a diagram

"Show me the diagram rules before I edit assembly-diagram.xml"

get_patterns serves the curated knowledge base — the cross-integration rules that span step types (EMF @mixed index math, swimlane layout, the three-entry add/remove rule, MVEL/XSLT/RAAS idioms). Call it with no args for an index, topic="Diagram Rules" for a section, or search="swimlane" to find by keyword. These are the same lessons in docs/studio-integration-patterns.md, now reachable from any MCP session.

Capture a discovery

When Claude hits an undocumented Studio behavior during a session, it logs the pattern to learnings.md via log_learning — entries are reviewed and promoted into the curated knowledge base.

Parse a server log

After downloading a server-{wid}.log from Workday (View Integration Events → expand documents → click the server-*.log):

"Parse my latest server log"

Returns structured timeline, unique errors, and XSLT messages. The parser auto-finds the most recent server-*.log in ~/Downloads.


Project structure

Workday-studio-mcp/
├── src/
│   ├── index.mjs               # Entry — registers tools, starts server
│   ├── config.mjs              # Loads workspace path
│   ├── fs.mjs                  # FS helpers + path traversal protection
│   ├── xml.mjs                 # XML validation wrapper
│   ├── assembly-validator.mjs  # Studio-specific assembly rules
│   ├── aidlc-docs.mjs          # Persists the integration plan + lifecycle state
│   └── tools/
│       ├── list-projects.mjs
│       ├── list-files.mjs
│       ├── read-file.mjs
│       ├── write-file.mjs
│       ├── search-files.mjs
│       ├── workspace-tree.mjs
│       ├── validate-xml.mjs
│       ├── create-project.mjs
│       ├── list-assembly-steps.mjs
│       ├── list-integration-params.mjs
│       ├── add-assembly-step.mjs
│       ├── create-xsl-transform.mjs
│       ├── copy-file-from-project.mjs
│       ├── rename-file.mjs
│       ├── delete-file.mjs
│       ├── get-step-type-reference.mjs   # Step type docs
│       ├── lookup-soap-operations.mjs    # WWS service/operation catalog
│       ├── get-patterns.mjs              # Serves the curated knowledge base
│       ├── plan-integration.mjs          # Design elicitation
│       ├── get-workflow.mjs              # Lifecycle status + next action
│       ├── update-sub-flow.mjs           # Surgical sub-flow replacement
│       ├── rename-steps.mjs              # Atomic step rename (assembly + diagram)
│       ├── delete-assembly-step.mjs      # Atomic step delete (assembly + diagram)
│       ├── validate-assembly.mjs         # Studio rules engine
│       ├── log-learning.mjs              # Knowledge-capture intake
│       └── parse-server-log.mjs          # Local log parser
├── docs/
│   └── studio-integration-patterns.md    # Shared knowledge base
├── learnings.md                # Append-only learnings intake queue
├── bin/
│   ├── install.sh
│   └── quickstart.sh
├── test/                       # `npm test` — node:test, no dependencies
│   ├── fixtures/
│   └── *.test.mjs
├── config.json.example
├── CLAUDE.md                   # Instructions for Claude when working in this repo
├── package.json
└── .gitignore

Security

  • The server only sees files inside your configured workspace_path — path traversal attempts are blocked.
  • No credentials, API keys, or Workday tenant details are stored or transmitted.
  • config.json (which contains your local workspace path) is gitignored.
  • The server runs over stdio — no network ports are opened.

Contributing patterns back

When you discover a new Studio quirk, schema rule, or assembly pattern, add it to docs/studio-integration-patterns.md and open a PR. See CLAUDE.md for guidance on what kinds of learnings belong there. The goal is a collective memory across the team — every debugging session that finds a new gotcha makes the next one cheaper.


Troubleshooting

workspace_path not configured Run cp config.json.example config.json and set the path.

Workspace path does not exist Check the path in config.json matches your Studio workspace folder.

Tools don't appear in Claude Ensure the path in your Claude config is absolute, not relative. Restart Claude Desktop after editing.

node: command not found in Claude Use the full node path:

/usr/local/bin/node /full/path/to/src/index.mjs

Find it with which node.


License

MIT

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
E2B

E2B

Using MCP to run code via e2b.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured