Official agent skill

Sop Builder

by IBM in IBM/ibm-watsonx-orchestrate-adk

Expert guidance for building a Standard Operating Procedure (SOP) from a workflow diagram, Langflow JSON, n8n JSON, BPMN model or workflow description.

OfficialMITAuto-check passedBusiness, Finance & HR

Install Sop Builder

skills CLI
$ npx skills add IBM/ibm-watsonx-orchestrate-adk --skill sop-builder -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install IBM/ibm-watsonx-orchestrate-adk sop-builder --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/IBM/ibm-watsonx-orchestrate-adk.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/sop-builder .claude/skills/sop-builder && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
sop-builder
GitHub stars
178
Token cost
~6.6k tokens
SKILL.md length
1,929 words
Files
1
Skills in repo
8
Repo updated
First seen
Licence
MIT

At a glance

Expert guidance for building a Standard Operating Procedure (SOP) from a workflow diagram, Langflow JSON, n8n JSON, BPMN model or workflow description.

  • Works in 12 steps: Executive Summary → Business Process Flow Diagram → Business Context → …
  • Tasks that involve Operations and SOPs
  • SKILL.md covers Purpose, Objective, Documentation Principles and Scope, plus 4 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Sop Builder is an agent skill from IBM/ibm-watsonx-orchestrate-adk, published by the product's own GitHub organization. Expert guidance for building a Standard Operating Procedure (SOP) from a workflow diagram, Langflow JSON, n8n JSON, BPMN model or workflow description. Focuses on business problem, data requirements, and business logic in plain language.

Its SKILL.md is about 6.6k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Business, Finance & HR, covering Operations and SOPs and Workflow automation. It works with n8n. The repository describes itself as: The command line client for watsonx Orchestrate's agent builder experience. The licence is MIT.

When your agent uses it

  • Tasks that involve Operations and SOPs
  • Tasks that involve Workflow automation

Example prompts

  • “/sop-builder”

Requirements

  • Python 3

Workflow steps

12 steps, taken from the step headings in SKILL.md.

  1. Executive Summary
  2. Business Process Flow Diagram
  3. Business Context
  4. Procedure Overview
  5. Data Requirements
  6. Custom Logic Documentation
  7. Identify Process Flow from Edges
  8. Extract Business Logic from Nodes
  9. Understand Data Flow
  10. Identify Decision Points
  11. Extract Business Rules
  12. Document Integration Points

What it can do on your machine

Read from SKILL.md and the folder at commit b6f9065. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are mermaid).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Sop Builder loads about 6.6k tokens when it runs. Until then it costs about 62 tokens; SKILL.md has 1,929 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~62
When it runs · the whole SKILL.md, loaded when a task matches
~6.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from IBM/ibm-watsonx-orchestrate-adk at commit b6f9065, republished under its MIT licence (© IBM). 1,929 words, ~6,621 tokens.

Download SKILL.mdSave it as .claude/skills/sop-builder/SKILL.md (or your agent's skills folder).
name
sop-builder
description
Expert guidance for building a Standard Operating Procedure (SOP) from a workflow diagram, Langflow JSON, n8n JSON, BPMN model or workflow description. Focuses on business problem, data requirements, and business logic in plain language.

Standard Operating Procedure (SOP) Generation

Purpose

This SOP defines the process for analyzing BPMN diagrams, Langflow JSON specifications, or n8n workflow descriptions and generating SOPs that describe the business problem, data requirements, and business logic in plain language—without technical BPMN, Langflow, or workflow automation terminology.

Objective

Transform technical workflow diagrams into SOP documentation that:

  • Can be understood by non-technical stakeholders
  • Focuses on business problems and solutions
  • Defines data requirements clearly
  • Describes business logic and decision-making
  • Serves as a bridge between business requirements and technical implementation
  • Contains ONLY factual information from the source specification

Documentation Principles

IMPORTANT: Generate specifications based ONLY on factual information present in the source workflow/diagram.

Rules:

  • Include ONLY what is explicitly defined or directly observable in the source
  • Do NOT add speculative content about performance, timing, or optimization
  • Do NOT include assumptions about non-functional characteristics unless explicitly specified
  • Do NOT add arbitrary concerns, compliance requirements, or improvement suggestions
  • Do NOT estimate durations, volumes, KPIs, or SLAs unless they are defined in the source
  • Focus on WHAT the process does, not HOW WELL it might perform
  • State when elements are "Not explicitly defined" rather than making assumptions

Exception - Technical Dependencies:

  • DO document any APIs, MCP servers, security protocols, storage systems, databases, or other technical dependencies that are explicitly referenced in the specification
  • Include specific details about these dependencies as stated in the source (e.g., API endpoints, authentication methods, database types, storage locations, MCP server configurations)
  • Document integration requirements and technical constraints that are specified in the source

Scope

This procedure applies to:

  • BPMN diagrams (XML or visual)
  • Langflow JSON specifications (exported flows)
  • n8n workflows
  • Other workflow automation specifications
  • Business process descriptions

The output is an SOP written in natural language for business stakeholders.

Langflow JSON Structure

When analyzing Langflow JSON files, extract information from:

  • Flow metadata: Name, description, and overall purpose
  • Nodes: Individual components representing process steps
    • ChatInput: Entry points for user input
    • ChatOutput: Exit points for results
    • Custom components: Business logic implementations
    • LLM components: AI-powered decision making
    • Data processing components: Transformations and validations
  • Edges: Connections between nodes showing process flow
  • Global variables: Configuration and dynamic values
  • Component templates: Detailed configuration of each node including:
    • Input/output specifications
    • Field configurations
    • Validation rules
    • Business logic embedded in code fields

Procedure

You are a business analyst specializing in translating technical workflows into clear business procedures. When provided with a workflow source (BPMN, n8n, or description), generate an SOP following this structure.

1. Executive Summary

Provide a concise overview for executives and decision-makers:

  • Procedure Name: Clear, business-friendly name
  • Business Problem: What problem does this procedure solve?
  • Business Objective: What business outcome does this achieve?
  • Scope: What is included and excluded
  • Key Benefits: Top 3-5 business benefits
  • Stakeholders: Who is involved or affected
  • Success Criteria: How do we know this procedure is working?
2. Business Process Flow Diagram

Create a Mermaid flowchart diagram that visualizes the business process flow:

Requirements:

  • Use Mermaid flowchart syntax (mermaid flowchart TD)
  • Show all major process steps as rectangles
  • Show decision points as diamonds
  • Show start and end points clearly
  • Use arrows to show flow direction
  • Include conditional branching where applicable
  • Use color coding to highlight different types of activities:
    • Start/End points (green/red)
    • Decision points (yellow)
    • Key business activities (blue for critical steps)
    • Standard process steps (white/default)
  • Keep labels concise but descriptive
  • Show loops for iterative processes

Include a Legend explaining:

  • What each shape represents
  • What colors indicate
  • Key flow characteristics (sequential, parallel, conditional, etc.)
  • Any special notations used

Example Structure:

mermaid
flowchart TD
    Start([Start Event]) --> Step1[Process Step]
    Step1 --> Decision{Decision Point?}
    Decision -->|Yes| Step2[Action A]
    Decision -->|No| Step3[Action B]
    Step2 --> End([End])
    Step3 --> End
3. Business Context
3.1 Problem Statement

Describe the business problem in detail:

  • What challenge or inefficiency exists today?
  • What is the impact of not solving this problem?
  • Who is affected by this problem?
  • What triggers the need for this procedure?
3.2 Current State (If Applicable)
  • How is this process handled today?
  • What are the pain points?
  • What manual steps exist?
  • What inefficiencies or risks are present?
3.3 Desired Future State
  • What will change with this procedure?
  • What improvements will be realized?
  • What new capabilities will be enabled?
4. Procedure Overview
4.1 Purpose and Scope
  • Primary Purpose: What this procedure accomplishes
  • Secondary Purposes: Additional benefits or outcomes
  • In Scope: What activities are included
  • Out of Scope: What is explicitly not included
  • Dependencies: What must exist or happen first
4.2 Roles and Responsibilities

For each role involved:

  • Role Name: (e.g., Sales Representative, Customer Service Agent)
  • Responsibilities: What they do in this procedure
  • Decision Authority: What decisions they can make
  • Escalation Path: When and to whom they escalate
4.3 Frequency and Timing
  • Trigger: What initiates this procedure
  • Frequency: How often it runs (daily, on-demand, etc.)
  • Duration: Expected time to complete
  • Business Hours: When this procedure operates
  • Urgency: Priority level and time sensitivity
5. Data Requirements
5.1 Input Data

For each data element needed to start the procedure:

Data Element Name:

  • Description: What this data represents
  • Source: Where it comes from
  • Format: How it's structured (in business terms)
  • Required/Optional: Is it mandatory?
  • Validation Rules: What makes it valid?
  • Example: Sample data for clarity

Example:

Customer Order Information:
- Description: Details about what the customer wants to purchase
- Source: Customer via website, phone, or email
- Format: Order form with customer details, items, quantities, and delivery address
- Required: Yes
- Validation Rules: Must include valid customer contact, at least one item, and delivery address
- Example: John Smith orders 2 widgets to be delivered to 123 Main St
5.2 Data Used During Procedure

Document data that is accessed or checked:

Data Element Name:

  • Description: What this data represents
  • Purpose: Why we need it
  • Source System: Where it's stored
  • Access Frequency: How often it's checked
  • Update Frequency: How often it changes
5.3 Output Data

For each data element produced:

Data Element Name:

  • Description: What this data represents
  • Purpose: How it will be used
  • Destination: Where it goes
  • Format: How it's structured
  • Retention: How long it's kept
6. Custom Logic Documentation

If the specification contains custom logic implemented in programming languages (JavaScript, Java, Python, etc.), document this logic in pseudo Python code format for clarity and consistency.

For each custom logic block:

Logic [Number]: [Logic Name/Purpose]

  • Original Language: The programming language used in the source
  • Purpose: What this logic accomplishes in business terms
  • Inputs: What data/variables are used
  • Outputs: What is produced or returned
  • Pseudo Python Code: Simplified Python-like representation of the logic

Guidelines for Pseudo Python Code:

  • Use clear, readable Python syntax
  • Add comments to explain business logic
  • Simplify complex operations while preserving intent
  • Use descriptive variable names
  • Include error handling if present in original
  • Show conditional logic clearly
  • Document any external function calls or API interactions

Example:

Logic 1: Calculate Order Discount
- Original Language: JavaScript
- Purpose: Determine the discount percentage based on order total and customer loyalty tier
- Inputs: order_total (number), customer_tier (string)
- Outputs: discount_percentage (number)
- Pseudo Python Code:
```python
def calculate_order_discount(order_total, customer_tier):
    """
    Calculate discount based on order value and customer loyalty tier.
    Business Rule: Higher tiers get better discounts, larger orders get additional discounts.
    """
    # Base discount by customer tier
    if customer_tier == "platinum":
        base_discount = 0.15  # 15% for platinum customers
    elif customer_tier == "gold":
        base_discount = 0.10  # 10% for gold customers
    elif customer_tier == "silver":
        base_discount = 0.05  # 5% for silver customers
    else:
        base_discount = 0.0   # No discount for regular customers
    
    # Additional discount for large orders
    if order_total > 1000:
        volume_discount = 0.05  # Extra 5% for orders over $1000
    elif order_total > 500:
        volume_discount = 0.02  # Extra 2% for orders over $500
    else:
        volume_discount = 0.0
    
    # Calculate total discount (capped at 25%)
    total_discount = min(base_discount + volume_discount, 0.25)
    
    return total_discount


### 6.1 LLM Prompts Documentation

If the specification contains LLM system prompts or user prompts, document them in this section.

**For each LLM prompt:**

**Prompt [Number]: [Prompt Name/Purpose]**
- **Prompt Type**: System Prompt or User Prompt
- **Purpose**: What this prompt accomplishes in business terms
- **Context**: When and where this prompt is used in the process
- **Prompt Content**: The actual prompt text (preserve formatting and structure)
- **Expected Output**: What type of response or behavior is expected
- **Variables/Placeholders**: Any dynamic elements that get substituted

**Example:**

Prompt 1: Customer Service Response Generator

  • Prompt Type: System Prompt

  • Purpose: Guide the AI to generate professional, empathetic customer service responses

  • Context: Used when responding to customer inquiries or complaints

  • Prompt Content: """ You are a professional customer service representative for [Company Name]. Your role is to:

    • Respond with empathy and understanding
    • Provide clear, actionable solutions
    • Maintain a friendly but professional tone
    • Escalate complex issues when appropriate

    Customer Context: {customer_tier}, {previous_interactions} Issue Type: {issue_category} """

  • Expected Output: A customer service response that addresses the issue while maintaining brand voice

  • Variables/Placeholders:

    • {customer_tier}: Customer loyalty level (bronze/silver/gold/platinum)
    • {previous_interactions}: Number of previous support tickets
    • {issue_category}: Type of issue (billing, technical, general inquiry)

**Note**: Only include this section if LLM prompts exist in the source specification. If no prompts are present, omit this section entirely.

### 7. Business Procedure Steps

Write the procedure as a numbered list of business steps (not technical tasks). Use clear, action-oriented language.

**Format for each step:**

**Step [Number]: [Action Name]**
- **What Happens**: Describe the business activity in plain language
- **Who Does It**: Role responsible (or "System" if automated)
- **Why**: Business reason for this step
- **Inputs**: What information is needed
- **Outputs**: What is produced or updated
- **Duration**: Typical time required
- **Success Criteria**: How to know it's done correctly

**Example:**

Step 1: Verify Customer Order

  • What Happens: Review the customer's order to ensure all required information is complete and accurate
  • Who Does It: Order Processing Team (or System if automated)
  • Why: Prevent processing incomplete or incorrect orders that would cause delays or errors
  • Inputs: Customer order form with items, quantities, delivery address, and payment method
  • Outputs: Validated order marked as "Ready for Processing" or "Needs Correction"
  • Duration: 2-5 minutes
  • Success Criteria: All required fields are complete, address is valid, items are available in catalog

### 8. Decision Points

For each decision in the procedure:

**Decision [Number]: [Decision Name]**
- **Question Being Answered**: What needs to be decided?
- **Who Decides**: Role or system making the decision
- **Decision Criteria**: What factors are considered?
- **Possible Outcomes**: What are the options?
- **Business Rules**: Rules that govern this decision
- **Impact**: What happens based on each outcome?

**Example:**

Decision 1: Is Inventory Available?

  • Question Being Answered: Do we have enough stock to fulfill this order?
  • Who Decides: Inventory Management System
  • Decision Criteria: Current stock levels vs. order quantity for each item
  • Possible Outcomes:
    • Yes - All items in stock → Proceed to payment
    • No - One or more items out of stock → Cancel order and notify customer
  • Business Rules:
    • Must have 100% of items available (no partial fulfillment)
    • Stock check must be real-time to prevent overselling
  • Impact:
    • If Yes: Customer gets their order, revenue is captured
    • If No: Customer is disappointed but informed, no false promises

### 9. Business Rules

Document all business rules that govern this procedure:

**Rule [Number]: [Rule Name]**
- **Rule Statement**: Clear statement of the rule
- **Business Rationale**: Why this rule exists
- **Applies To**: What situations or steps
- **Enforced By**: Who or what enforces it
- **Exceptions**: When this rule doesn't apply
- **Consequences**: What happens if violated

**Example:**

Rule 1: Payment Before Shipment

  • Rule Statement: No order shall be shipped until payment is successfully processed and confirmed
  • Business Rationale: Protect company revenue and minimize financial risk from unpaid orders
  • Applies To: All customer orders regardless of size or customer type
  • Enforced By: Order Management System (automated control)
  • Exceptions: Pre-approved corporate accounts with net-30 terms
  • Consequences: If violated, company loses revenue and incurs shipping costs for unpaid orders

### 10. Exception Handling

Document what happens when things go wrong:

**Exception [Number]: [Exception Name]**
- **What Goes Wrong**: Description of the problem
- **How It's Detected**: How we know there's a problem
- **Business Impact**: What's at risk
- **Response Procedure**: What we do about it
- **Responsible Party**: Who handles it
- **Prevention**: How to avoid this in the future
- **Escalation**: When to escalate and to whom

### 10. Integration Points

Describe interactions with other systems or procedures (in business terms):

**Integration [Number]: [System/Procedure Name]**
- **Purpose**: Why we interact with this system
- **What We Send**: Information provided to the system
- **What We Receive**: Information received back
- **Timing**: When this interaction happens
- **Dependency**: What happens if this system is unavailable
- **Business Owner**: Who owns the other system/procedure

### 11. Notes and Observations

#### 11.1 Process Characteristics
Document observable characteristics from the source specification:
- Process complexity (simple, moderate, complex)
- Automation level (manual, semi-automated, fully automated)
- Integration requirements (systems involved)
- Data volume characteristics (if specified)

#### 11.2 Limitations and Constraints
Document only what is explicitly stated or observable:
- Known limitations in the current design
- Explicit constraints mentioned in the specification
- Dependencies that must be satisfied
- Scope boundaries

#### 11.3 Additional Context
Include only if explicitly provided in the source:
- Regulatory requirements (if specified)
- Compliance needs (if specified)
- Quality standards (if specified)
- Performance targets (if specified)

**Note**: If any of these elements are not defined in the source specification, state "Not explicitly defined in source specification" rather than making assumptions.

---

## Writing Guidelines

### Language and Style
- **Use Business Language**: Avoid technical jargon, BPMN terms, or automation terminology
- **Be Specific**: Use concrete examples and scenarios
- **Be Action-Oriented**: Focus on what happens and why, not how it's implemented
- **Be Clear**: Write for someone unfamiliar with the process
- **Be Concise**: Eliminate unnecessary words while maintaining clarity

### Terminology to Avoid
❌ **Don't Use**: BPMN, nodes, gateways, sequence flows, service tasks, message events, XML, JSON, Langflow components, edges, ChatInput, ChatOutput, API calls, webhooks, HTTP requests, database queries, global variables, component templates

✅ **Do Use**: Steps, decisions, activities, information, data, systems, checks, notifications, updates, records, forms, reports, inputs, outputs, configuration values

### Examples of Translation

**BPMN Technical**: "The exclusive gateway evaluates the inventory_available boolean attribute"
**Business**: "We check whether we have enough stock to fulfill the order"

**BPMN Technical**: "A service task calls the payment API endpoint with order data"
**Business**: "We process the customer's payment through our payment system"

**BPMN Technical**: "The subprocess contains embedded error boundary events"
**Business**: "If something goes wrong during this step, we have procedures to handle the problem"

**Langflow Technical**: "The ChatInput node receives user message and passes it to the ActionItemExtractor component"
**Business**: "The system receives the user's request and analyzes it to identify action items"

**Langflow Technical**: "The component template defines input_types as ['Message', 'Data', 'Text']"
**Business**: "The system accepts information in various formats including messages and structured data"

**Langflow Technical**: "The edge connects node A's output handle to node B's input handle"
**Business**: "Information flows from step A to step B"

**Langflow Technical**: "Global variable 'api_key' is configured with type 'str' and marked as required"
**Business**: "The system requires authentication credentials to access external services"

### Formatting Standards
- Use clear headings and subheadings
- Use bullet points for lists
- Use numbered lists for sequential steps
- Use tables for structured comparisons
- Use examples to illustrate complex concepts
- Use bold for emphasis on key terms
- Use quotes for specific data values or examples

---

## Quality Standards

### Completeness
- All sections must be addressed
- Sections without source information should state "Not explicitly defined in source specification"
- All decisions must have clear criteria from the source
- All data elements must be defined based on source information

### Accuracy
- Information must match the source workflow exactly
- No hallucination or invention of details
- Do NOT make assumptions - only document what is observable
- If information is inferred, clearly state "Inferred from [specific source element]"
- If information is missing, state "Not explicitly defined in source specification"

### Clarity
- Readable by non-technical business stakeholders
- Free of technical jargon
- Concrete examples provided where available in source
- Logical flow and organization

### Business Focus
- Emphasizes business value and outcomes observable in the workflow
- Explains "why" based on what can be deduced from the source
- Connects to business objectives only when evident in the source
- Addresses stakeholder concerns only when specified in the source

---

## Output Format

### File Naming Convention
`[procedure-name]-sop.md`

Example: `order-fulfillment-sop.md`

### Document Header
```markdown
# [Procedure Name] - SOP

**Document Type**: SOP
**Version**: 1.0
**Date**: [Date]
**Author**: [Name]
**Source**: [Original workflow file/diagram]
**Status**: [Draft/Review/Approved]

---

## Document Control

**Approvers**:
- Business Owner: [Name/TBD]
- Process Owner: [Name/TBD]
- Compliance: [Name/TBD]

**Review Cycle**: [Frequency]
**Next Review Date**: [Date]

---
Show full SKILL.md (552 more words)Show less
Document Structure

Follow the sections outlined above in order:

  1. Executive Summary
  2. Business Process Flow Diagram
  3. Business Context
  4. Procedure Overview
  5. Data Requirements
  6. Custom Logic Documentation (if applicable) 6.1. LLM Prompts Documentation (if applicable)
  7. Business Procedure Steps
  8. Decision Points
  9. Business Rules
  10. Exception Handling
  11. Integration Points
  12. Notes and Observations

Usage Instructions

To Use This SOP:
  1. Provide the Source: Share the BPMN diagram, Langflow JSON file, n8n workflow, or process description
  2. Specify Context: Provide any additional business context or constraints
  3. Review Output: Verify the business specification for accuracy
  4. Iterate: Request clarifications or additional details as needed
  5. Validate: Have business stakeholders review and approve
  6. Maintain: Update as the procedure evolves
Example Request (BPMN):
Please analyze the following workflow and generate an SOP:

@order-fulfillment-process.bpmn

Additional context:
- This is for an e-commerce company
- Average order value is $150
- Processing 500-1000 orders per day
- Customer service team has 10 people
- Focus on customer experience and efficiency
Example Request (Langflow):
Please analyze the following Langflow workflow and generate an SOP:

@action-item-extractor.json

Additional context:
- This is for a project management team
- Processes meeting transcripts and chat logs
- Extracts actionable tasks with assignees and due dates
- Used by 50+ team members
- Focus on task tracking and accountability

Provide Your Workflow Below:

Workflow Source:

@your-workflow-file  # Replace with your BPMN, Langflow JSON, n8n, or workflow file

Additional Context (optional but recommended):

- Industry/business type:
- Scale/volume:
- Key stakeholders:
- Primary objectives:
- Known constraints:
- Specific concerns:

Workflow-Specific Extraction Guidelines

Langflow JSON Files

When analyzing Langflow JSON files, follow these additional guidelines:

1. Identify Process Flow from Edges
  • Map edges array to understand the sequence of operations
  • source and target fields indicate flow direction
  • Multiple edges from one node indicate branching/decision points
  • Edges returning to previous nodes indicate loops
2. Extract Business Logic from Nodes

For each node in the nodes array:

  • Node ID: Used for tracking but translate to business step names
  • Node Type: Indicates the component type (ChatInput, ChatOutput, custom components)
  • Display Name: Often contains business-friendly names
  • Description: May contain business purpose
  • Template Fields: Configuration that reveals business rules and logic
  • Code Fields: May contain embedded business logic (translate to pseudo-code)
3. Understand Data Flow
  • Input/Output Types: Reveal what data moves between steps
  • Field Configurations: Show validation rules and data requirements
  • Global Variables: Indicate external dependencies and configuration needs
4. Identify Decision Points

Look for:

  • Conditional components
  • Multiple outgoing edges from a single node
  • Components with boolean outputs
  • Branching logic in custom code
5. Extract Business Rules

From:

  • Validation rules in field configurations
  • Conditional logic in custom components
  • Required vs optional field settings
  • Data type constraints
6. Document Integration Points

Identify from:

  • API call components
  • Database query components
  • External service integrations
  • Global variables pointing to external systems
n8n Workflows

When analyzing n8n workflow JSON files, follow these additional guidelines:

1. Identify Process Flow from Connections
  • Map connections object to understand the sequence of operations
  • Each node's connections show which nodes receive its output
  • Multiple connections from one node indicate branching/decision points
  • Connections returning to previous nodes indicate loops
2. Extract Business Logic from Nodes

For each node in the nodes array:

  • Node Name: Translate to business step names
  • Node Type: Indicates the operation type (HTTP Request, IF, Set, etc.)
  • Parameters: Configuration that reveals business rules and logic
  • Notes: May contain business purpose and context
  • Credentials: Indicate external system integrations
3. Understand Data Flow
  • Node Outputs: Reveal what data moves between steps
  • Expressions: Show data transformations and calculations
  • Parameters: Show validation rules and data requirements
4. Identify Decision Points

Look for:

  • IF nodes for conditional logic
  • Switch nodes for multi-way branching
  • Filter nodes for data-based routing
  • Multiple output connections from a single node
5. Extract Business Rules

From:

  • IF node conditions
  • Filter node criteria
  • Set node value mappings
  • Function node logic (translate to pseudo-code)
6. Document Integration Points

Identify from:

  • HTTP Request nodes (API calls)
  • Database nodes (queries and updates)
  • Webhook nodes (external triggers)
  • Credential configurations (external systems)

© IBM, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/sop-builder of IBM/ibm-watsonx-orchestrate-adk.

Open the folder on GitHubat commit b6f9065

Compare with similar skills

Sop Builder next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Sop Builder compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Sop Builder this skillIBM/ibm-watsonx-orchestrate-adk178—~6.6kAutomated safety check: PassMIT
N8n Add Workflowkhanhduyvt0101/workflows110—~1.3kAutomated safety check: PassNone
Airtable Automationclaude-office-skills/skills498—~2.7kAutomated safety check: PassMIT
N8n MCP Tools Expertdavila7/claude-code-templates32k8 repos~3.2kAutomated safety check: PassMIT
Automation InstructionsdeusXmachina-dev/memorylane121—~1.1kAutomated safety check: PassGPL-3.0
n8n Expression Syntaxczlonkowski/n8n-skills6.4k—~5.2kAutomated safety check: PassMIT

Similar skills

  • N8n Add Workflow

    khanhduyvt0101/workflows

    Add a new n8n workflow template to the repository. An agent skill from khanhduyvt0101/workflows.

    110 GitHub stars~1.3k tokensUpdated 4 mo ago
    Productivity & AutomationAuto-check passed
  • Airtable Automation

    claude-office-skills/skills

    Airtable database automation - views, automations, integrations, and workflow triggers

    498 GitHub stars~2.7k tokensUpdated 8 mo ago
    Productivity & AutomationAuto-check passed
  • N8n MCP Tools Expert

    davila7/claude-code-templates

    Expert guide for using n8n-mcp MCP tools effectively. An agent skill from davila7/claude-code-templates.

    32k GitHub starsUsed in 8 repos~3.2k tokens
    Productivity & AutomationAuto-check passed
  • Automation Instructions

    deusXmachina-dev/memorylane

    Write step-by-step automation instructions for a workflow, tailored to your tool (Claude, n8n or Zapier).

    121 GitHub stars~1.1k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • n8n Expression Syntax

    czlonkowski/n8n-skills

    Validates and fixes n8n's double-curly-brace expressions, covering $json, $node, and the webhook body-wrapping mistake.

    6.4k GitHub stars~5.2k tokensUpdated 21 days ago
    Productivity & AutomationAuto-check passed
  • N8n Code Javascript

    davila7/claude-code-templates

    Write JavaScript code in n8n Code nodes. An agent skill from davila7/claude-code-templates.

    32k GitHub starsUsed in 4 repos~4k tokens
    Productivity & AutomationAuto-check passed

More from IBM/ibm-watsonx-orchestrate-adk

All 8 skills in this repo
  • Agentic Workflow Advisor

    IBM/ibm-watsonx-orchestrate-adk

    Official

    Analyzes IBM watsonx Orchestrate agentic workflow artefacts (JSON or Python @flow) and returns prioritised architecture recommendations grouped by impact.

    178 GitHub stars~10k tokensUpdated 4 days ago
    Auto-check passed
  • Telemetry Analyzer

    IBM/ibm-watsonx-orchestrate-adk

    Official

    A skill your agent uses when the user wants to analyze agent telemetry traces to find bugs and get fix recommendations — walks through exporting traces from a local or remote watsonx Orchestrate…

    178 GitHub stars~10k tokensUpdated 4 days ago
    Auto-check: notes
  • Customercare MCP Builder

    IBM/ibm-watsonx-orchestrate-adk

    Official

    Build MCP servers for customer care agents following Watson Orchestrate specifications.

    178 GitHub stars~5.3k tokensUpdated 4 days ago
    Auto-check passed
  • Wxo Builder

    IBM/ibm-watsonx-orchestrate-adk

    Official

    A skill your agent uses when building, testing, debugging, or publishing IBM watsonx Orchestrate agents, tools, flows, connections, knowledge bases, or custom models with the orchestrate CLI or ADK…

    178 GitHub stars~6.8k tokensUpdated 4 days ago
    Auto-check: notes
  • Agent Instructions Evaluator

    IBM/ibm-watsonx-orchestrate-adk

    Official

    Evaluate an agent instructions or agent definition for achievability and produce a structured, evidence-backed report artifact with per-dimension scores, findings, deterministic signals, and…

    178 GitHub stars~17k tokensUpdated 4 days ago
    Auto-check passed
  • Solution Architect

    IBM/ibm-watsonx-orchestrate-adk

    Official

    Expert guidance for creating high-level solution architecture documents from business requirements, use cases, or problem statements.

    178 GitHub stars~8.4k tokensUpdated 4 days ago
    Auto-check passed

Works with

Questions about Sop Builder

What does Sop Builder do?

Expert guidance for building a Standard Operating Procedure (SOP) from a workflow diagram, Langflow JSON, n8n JSON, BPMN model or workflow description. Sop Builder is an agent skill from IBM/ibm-watsonx-orchestrate-adk, published by the product's own GitHub organization. Expert guidance for building a Standard Operating Procedure (SOP) from a workflow diagram, Langflow JSON, n8n JSON, BPMN model or workflow description.

When should I use Sop Builder?

Sop Builder fits situations like: tasks that involve Operations and SOPs; tasks that involve Workflow automation.

How do I install Sop Builder in Claude Code?

Run `npx skills add IBM/ibm-watsonx-orchestrate-adk --skill sop-builder -a claude-code`. Or copy the skill folder (skills/sop-builder in IBM/ibm-watsonx-orchestrate-adk) into .claude/skills/sop-builder in your project. Claude Code loads it when a task matches its description.

How do I install Sop Builder in Codex?

Run `npx skills add IBM/ibm-watsonx-orchestrate-adk --skill sop-builder -a codex`. Or copy the skill folder (skills/sop-builder in IBM/ibm-watsonx-orchestrate-adk) into .agents/skills/sop-builder in your project. Codex loads it when a task matches its description.

Can I use Sop Builder in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add IBM/ibm-watsonx-orchestrate-adk --skill sop-builder -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/sop-builder, .gemini/skills/sop-builder, .github/skills/sop-builder and .opencode/skills/sop-builder in your project.

What does Sop Builder need to run?

SKILL.md names no scripts, command-line tools or credentials: Sop Builder is instructions for the agent only. Our summary lists: Python 3.

Does Sop Builder access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Sop Builder safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Sop Builder use?

Sop Builder is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Sop Builder use?

About 6.6k tokens (SKILL.md is roughly 26k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Sop Builder?

Skills that share tags, products or a category with Sop Builder: N8n Add Workflow (khanhduyvt0101/workflows, 110 stars), Airtable Automation (claude-office-skills/skills, 498 stars), N8n MCP Tools Expert (davila7/claude-code-templates, 32k stars) and Automation Instructions (deusXmachina-dev/memorylane, 121 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Sop Builder?

IBM (a GitHub organization, an official publisher) maintains it in IBM/ibm-watsonx-orchestrate-adk, which has 178 GitHub stars. The repository holds 8 skills in this directory. The repository was last updated on October 2, 2026.

Source: IBM/ibm-watsonx-orchestrate-adk on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.