Overview
This skill set covers the complete journey of building enterprise AI agent solutions:- solution-architect - Transforms business problems into structured solution architectures with executive summaries, technical designs, and implementation roadmaps. Creates the foundation for SOP development.
- sop-builder - Converts technical procedures and workflows (e.g., BPMN, Langflow, n8n) into business-readable Standard Operating Procedures with plain-language descriptions, Mermaid diagrams, and detailed process documentation.
- wxo-builder - Implements production-ready watsonx Orchestrate solutions from SOPs or prompts, generating agents, tools, flows, and knowledge bases following ADK best practices and project structures.
- wxo-analyzer - Reverse-engineers existing wxO projects to produce comprehensive documentation, relationship diagrams, and anti-pattern analysis across enterprise AI failure modes for auditing and onboarding.
Available Skills
1. solution-architect
Location:solution-architect/SKILL.md
Focuses on turning business problems into clear, structured solution architectures. It bridges business and technical perspectives by producing:
- Executive-friendly solution overviews
- Technical architecture and design patterns
- Implementation roadmaps and SOP-ready documentation
- Analyzing business requirements and problem statements
- Generating three focused architecture documents (Solution Overview, Technical Architecture, Implementation Plan)
- Business-focused summaries and context
- Bridging business needs with technical implementation
- Preparing documentation for elaboration into detailed SOPs
- Modular, manageable documentation structure
2. sop-builder
Location:sop-builder/SKILL.md
Transforms technical workflows into business-readable SOPs. It:
- Converts BPMN, Langflow, and n8n workflows into plain-language procedures
- Documents decision points, exceptions, integrations, and prompts
- Produces visuals (Mermaid) and business-ready process descriptions
- Analyzing BPMN diagrams, Langflow JSON, n8n workflows
- Generating business-focused SOPs in plain language
- Business process flow diagrams with Mermaid
- Data requirements and custom logic documentation
- LLM prompts documentation
- Business procedure steps and decision points
- Exception handling and integration points
- Translation from technical to business language
3. wxo-builder
Location:wxo-builder/SKILL.md
Turns SOPs or prompts into fully implemented watsonx Orchestrate solutions. It:
- Generates agents, tools, flows, and knowledge bases
- Follows recommended ADK project structures and patterns
- Supports multiple KB backends and end-to-end CLI workflows
- Generating complete wxO implementations from Standard Operating Procedures (SOPs)
- Transforming business requirements into agents, tools, flows, and knowledge bases
- Recommended workflow: Use sop-builder to create SOPs from BPMN/n8n/Langflow first, then use wxo-builder to generate wxO solutions
- Knowledge base providers (Milvus, AstraDB, Elasticsearch)
- Standard project structure and implementation patterns
- Document processing flows and workflow patterns
- Python tool and flow decorators
- Agent YAML configuration
- CLI import commands and best practices
- Complete examples from the ADK repository
4. wxo-analyzer
Location:wxo-analyzer/SKILL.md
Reverse-engineers and documents existing wxO projects. It:
- Analyzes project structure and components
- Produces multi-report documentation (overview, agents, tools)
- Generates diagrams and relationship mappings
- Detects anti-patterns and provides recommendations
- Analyzing existing wxO project structures
- Generating three-report documentation sets (Solution Overview, Agent Analysis, Tools & Components)
- Creating Mermaid diagrams showing component relationships
- Documenting agents, tools, connections, and knowledge bases
- Project structure and component inventory
- Relationship mapping between agents, tools, and resources
- Anti-pattern detection across 15 enterprise AI failure modes
- Instruction line counting and complexity analysis
- Production readiness assessment
How They Fit Together
A typical end-to-end workflow looks like:- solution-architect establishes the foundation and high-level design
- sop-builder creates detailed, business-readable procedures
- wxo-builder implements the technical solution
- wxo-analyzer validates, documents, and identifies improvements
Using These Skills
In Claude Desktop or Web
- Navigate to the Skills section in Claude
- Import the desired skill by selecting the appropriate SKILL.md file:
solution-architect/SKILL.md- For solution architecture documentationsop-builder/SKILL.md- For SOP generation from workflowswxo-builder/SKILL.md- For watsonx Orchestrate developmentwxo-analyzer/SKILL.md- For analyzing existing wxO projects
- The skill will be available in your conversations
Agent Skills Overview
A standardized way to give AI agents new capabilities and expertise.What are Agent Skills?
Agent Skills are a lightweight, open format for extending AI agent capabilities with specialized knowledge and workflows. At its core, a skill is a folder containing a SKILL.md file. This file includes metadata (name and description, at minimum) and instructions that tell an agent how to perform a specific task. Skills can also bundle scripts, reference materials, templates, and other resources.Why Agent Skills?
Agents are increasingly capable, but often don’t have the context they need to do real work reliably. Skills solve this by packaging procedural knowledge and company-, team-, and user-specific context into portable, version-controlled folders that agents load on demand. This gives agents:- Domain expertise: Capture specialized knowledge — from legal review processes to data analysis pipelines to presentation formatting — as reusable instructions and resources.
- Repeatable workflows: Turn multi-step tasks into consistent, auditable procedures.
- Cross-product reuse: Build a skill once and use it across any skills-compatible agent.
How do Agent Skills work?
Agents load skills through progressive disclosure, in three stages:- Discovery: At startup, agents load only the name and description of each available skill, just enough to know when it might be relevant.
- Activation: When a task matches a skill’s description, the agent reads the full SKILL.md instructions into context.
- Execution: The agent follows the instructions, optionally executing bundled code or loading referenced files as needed.
Skill Structure
Each skill follows the Skills format:- SKILL.md: Main skill file with frontmatter (name, description)
- examples.md: Complete reference implementations and code examples (where applicable)
- Frontmatter: Contains skill metadata
- Content: Expert guidance, specifications, and best practices
Resources
- solution-architect:
solution-architect/SKILL.md - sop-builder:
sop-builder/SKILL.md - wxo-builder:
wxo-builder/SKILL.md - wxo-analyzer:
wxo-analyzer/SKILL.md - Reference Examples:
customercare-mcp-builder/references/examples.md - IBM watsonx Orchestrate ADK: https://github.com/IBM/ibm-watsonx-orchestrate-adk
- MCP Server:
packages/mcp-server/ibm_watsonx_orchestrate_mcp_server/

