skill-template

name	skill-template
description	Skeleton template for authoring quality engineering skills adhering to the agentskills.io standard.

Skill Template

Overview

Briefly describe the quality engineering domain, the core problem this skill addresses, and the governing quality standard references (e.g., AIAG & VDA FMEA 1st Edition, AIAG SPC 2nd Edition, ISO 9001:2015, IATF 16949:2016).

State the high-level objective of the skill and how it guides the AI agent through the domain-specific workflow.

When to Use

Specify the explicit triggers and scenarios for activating this skill:

• Trigger Scenarios: List the problem statements, user requests, or process events where this skill applies.
• Input Requirements: Detail required input artifacts (e.g., CSV datasets, process flow diagrams, BOMs, inspection logs, severity ratings).
• Prerequisites: Note any preceding steps, certifications, or tool connectivity requirements.

Step-by-Step Methodology

Provide the structured, phased engineering methodology:

1. Phase 1: Ingestion & Problem Decomposition
   • Collect and validate raw domain inputs.
   • Decompose complex systems into constituent functions, failure modes, or characteristic parameters.
2. Phase 2: Qualitative Engineering Reasoning
   • Apply standard quality heuristics (e.g., 6M classification, boundary definition).
   • Formulate structured payload objects for calculation engines.
3. Phase 3: Tool Execution & Deterministic Verification
   • Dispatch structured data to the designated quality-mcp tool.
   • Await deterministic results from the engine.
4. Phase 4: Synthesis & Actionable Recommendations
   • Interpret tool output in the context of industry standards.
   • Generate prioritized corrective actions, containment steps, or control plan updates.

Tool Invocations

Detail the exact MCP tools available for this skill.

tool_name

• MCP Server: quality-mcp
• Purpose: Deterministic calculation of domain metrics.
• Parameters:
  • param_1 (type): Description of parameter 1.
  • param_2 (type): Description of parameter 2.
• Expected Output:
  • result_field (type): Description of return field.
• Example Call:

  {
    "param_1": "value_1",
    "param_2": [1.0, 2.0, 3.0]
  }
  

Best Practices

• Strict Invariant: No Inline Math in Prompt Context. Never attempt to calculate statistical limits, capability indices, RPN/AP scores, or variance components directly in prompt text. Always delegate computation to quality-mcp tools.
• Standards Fidelity: Quote and align with official standards terminology (e.g., AIAG & VDA, ISO 9001).
• Evidence-Based Dispositions: Ensure all risk ratings, dispositions, and root cause conclusions are backed by objective evidence and deterministic tool outputs.
• Traceability: Maintain bidirectional linkage between failure modes, causes, controls, and verification data.