Update to BuyerTwin

{
  "description": "Use when the user asks for similarities, differences, patterns, or implications across two or more buyers based on canonical BuyerTwin intelligence.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 442
    },
    {
      "relative_path": "references/comparison-methodology.md",
      "size_in_bytes": 5249
    }
  ],
  "name": "compare-buyers",
  "skill_md_contents": "---\nname: compare-buyers\ndescription: Use when the user asks for similarities, differences, patterns, or implications across two or more buyers based on canonical BuyerTwin intelligence.\n---\n\nUse this skill when the user asks to compare buyers, such as shared pains, differing triggers, contrasting objections, universal themes, buyer-specific priorities, or where messaging should be universal vs personalized.\n\nDo not use this as the primary workflow for:\n- simulated buyer reactions to a new idea or change\n- role-play in buyer voice\n- content generation\n- messaging evaluation workflows\n- buyer knowledge mutation\n\nCore workflow:\n1. Resolve requested buyer scope deterministically before comparison.\n2. Retrieve canonical BuyerTwin intelligence relevant to the user question.\n3. Compare evidence across at least two buyers.\n4. Distinguish universal, majority, buyer-specific evidence, contrasting, and missing/unknown patterns only when evidence supports each label.\n5. Provide implications as AI synthesis while preserving evidence boundaries.\n\nBuyer scope rules:\n- Named buyer means resolve that specific buyer.\n- Multiple named buyers means resolve all named buyers.\n- Explicit segment or subset means compare that exact subset.\n- \"All buyers\" means all accessible buyers in scope.\n- \"My buyers\" without narrowing defaults to all accessible buyers in scope.\n- Do not silently narrow scope by model-defined relevance.\n- Do not silently omit buyers because they appear redundant.\n- If technical limits prevent full requested scope, disclose the limitation.\n- If fewer than two buyers are available, explain that reliable comparison is not possible.\n\nCanonical-first comparison rules:\n- Use canonical BuyerTwin intelligence as the evidence base.\n- Do not use ask_buyer as the normal comparison mechanism.\n- If canonical intelligence can answer directly, do not generate simulation-based restatements.\n- If evidence is missing, state what is missing instead of fabricating differences.\n\nEvidence hierarchy:\n- BuyerTwin Knowledge is authoritative for factual claims.\n- AI Synthesis may infer implications from that knowledge.\n- Do not present synthesis as a stored BuyerTwin fact.\n- Do not treat missing evidence as negative evidence.\n- Do not equate frequency with formal priority unless ranking evidence exists.\n\nFor detailed methodology and edge-case handling, consult:\n- references/comparison-methodology.md\n"
}