IBM Content Services MCP Server

Overview

The IBM Content Services MCP Server provides a standardized interface that enables IBM FileNet Content Manager (FNCM) capabilities to be used by AI models. The server is available in four specialized configurations to support different workflows:

• Core Server: Comprehensive document and content management operations
• Property Extraction and Classification Server: Document analysis for property extraction and classification
• Legal Hold Server: Legal hold management
• AI Document Insight Server: Uses virtual table capabilities for Content Assistant vector search combined with property/metadata search, along with document summarization, comparison, and Q&A

Each server can be deployed independently or in combination to support your specific use cases.

Server Types and Capabilities

Core Server (core-cs-mcp-server)

Purpose: Comprehensive document and content management operations

Use Cases:

• General document lifecycle management (create, update, check-in, check-out, delete)
• Folder operations and document filing
• Search and retrieval across the repository
• Metadata and class management
• Resources support for exposing documents as LLM context

Property Extraction and Classification Server (property-extraction-and-classification-cs-mcp-server)

Purpose: Document analysis for property extraction and classification

Dependencies: Requires Core Server for document update operations (e.g., update_document_properties)

Use Cases:

• Automated property extraction from document content
• Document classification and reclassification workflows
• Content-based metadata enrichment

Legal Hold Server (legal-hold-cs-mcp-server)

Purpose: Legal hold management

Use Cases:

• Creating and managing legal holds
• Placing documents and objects under hold
• Tracking held objects

AI Document Insight Server (ai-document-insight-cs-mcp-server)

Purpose: Uses virtual table capabilities for Content Assistant vector search combined with property/metadata search, along with document summarization, comparison, and Q&A

Dependencies:

• Requires Core Server for document operations
• Requires Persistent Text Extract add-on
• Requires Content Assistant add-on
• Requires FileNet 5.7.0 IF003 or later

Use Cases:

• Content-based document search with metadata filtering using Content Assistant API vector search
• Content Assistant API document summaries
• Content Assistant API document comparison and analysis

Tools List

Core Server Tools

The Core Server provides 26 tools organized into the following categories:

Document Management (11 tools)

• get_document_versions: Retrieves a document's version history, including major and minor version numbers and document IDs for each version.

• get_document_text_extract: Extracts text content from a document by retrieving its text extract annotations. If multiple text extracts are found, they are concatenated. IMPORTANT: This functionality requires the Persistent Text Extract add-on to be installed in your object store. See the Prerequisites section for more details.

• create_document: Creates a new document in the content repository with specified properties. Can upload files as the document's content if file paths are provided. Requires first calling determine_class and get_class_property_descriptions.

• update_document_properties: Updates an existing document's properties without changing its class. Requires first calling get_class_property_descriptions to get valid properties for the document's current class.

• update_document_class: Changes a document's class in the content repository. WARNING: Changing a document's class can result in loss of properties if the new class does not have the same properties as the old class. Requires first calling determine_class to get the new class_identifier.

• **checkin_do

... View full README on GitHub